<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Publishing DTD v1.1 20151215//EN" "https://jats.nlm.nih.gov/publishing/1.1/JATS-journalpublishing1.dtd">
<article article-type="research-article" dtd-version="1.1" specific-use="sps-1.9" xml:lang="en" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">
	<front>
		<journal-meta>
			<journal-id journal-id-type="nlm-ta">abchf</journal-id>
			<journal-id journal-id-type="publisher-id">abchf</journal-id>
			<journal-title-group>
				<journal-title>ABC Heart Fail Cardiomyop</journal-title>
				<abbrev-journal-title abbrev-type="publisher">ABC Heart Fail Cardiomyop</abbrev-journal-title>
			</journal-title-group>
			<issn pub-type="ppub">2764-3107</issn>
			<publisher>
				<publisher-name>Sociedade Brasileira de Cardiologia - SBC</publisher-name>
			</publisher>
		</journal-meta>
		<article-meta>
			<article-id pub-id-type="doi">10.36660/abchf.20260016i</article-id>
			<article-categories>
				<subj-group subj-group-type="heading">
					<subject>Original Article</subject>
				</subj-group>
			</article-categories>
			<title-group>
				<article-title>Global Mapping of Research on Artificial Intelligence Applied to Myocarditis</article-title>
			</title-group>
			<contrib-group>
				<contrib contrib-type="author">
					<contrib-id contrib-id-type="orcid">0000-0002-2386-048X</contrib-id>
					<name>
						<surname>Costa</surname>
						<given-names>Guilherme Cesar Fernandes de Oliveira da</given-names>
					</name>
					<role>Conception and design of the research and Statistical analysis</role>
					<role>Acquisition of data</role>
					<role>Analysis and interpretation of the data</role>
					<role>Writing of the manuscript and Critical revision of the manuscript for intellectual content</role>
					<xref ref-type="aff" rid="aff1"><sup>1</sup></xref>
				</contrib>
				<contrib contrib-type="author">
					<contrib-id contrib-id-type="orcid">0009-0004-7596-1014</contrib-id>
					<name>
						<surname>Rodriguez</surname>
						<given-names>Jose Gregorio Valero</given-names>
					</name>
					<role>Conception and design of the research and Statistical analysis</role>
					<role>Analysis and interpretation of the data</role>
					<role>Writing of the manuscript and Critical revision of the manuscript for intellectual content</role>
					<xref ref-type="aff" rid="aff1"><sup>1</sup></xref>
				</contrib>
				<contrib contrib-type="author">
					<contrib-id contrib-id-type="orcid">0009-0005-6008-4335</contrib-id>
					<name>
						<surname>Costa</surname>
						<given-names>Giulia Fernandes de Oliveira da</given-names>
					</name>
					<role>Conception and design of the research and Statistical analysis</role>
					<role>Acquisition of data</role>
					<role>Analysis and interpretation of the data</role>
					<role>Writing of the manuscript and Critical revision of the manuscript for intellectual content</role>
					<xref ref-type="aff" rid="aff2"><sup>2</sup></xref>
				</contrib>
				<contrib contrib-type="author">
					<contrib-id contrib-id-type="orcid">0000-0002-8870-5130</contrib-id>
					<name>
						<surname>Pimentel</surname>
						<given-names>Adriana Munford Lima</given-names>
					</name>
					<role>Analysis and interpretation of the data</role>
					<role>Writing of the manuscript and Critical revision of the manuscript for intellectual content</role>
					<xref ref-type="aff" rid="aff3"><sup>3</sup></xref>
				</contrib>
				<contrib contrib-type="author">
					<contrib-id contrib-id-type="orcid">0000-0002-7452-3870</contrib-id>
					<name>
						<surname>Mesquita</surname>
						<given-names>Evandro Tinoco</given-names>
					</name>
					<role>Analysis and interpretation of the data</role>
					<role>Writing of the manuscript and Critical revision of the manuscript for intellectual content</role>
					<xref ref-type="aff" rid="aff3"><sup>3</sup></xref>
				</contrib>
			</contrib-group>
			<aff id="aff1">
				<label>1</label>
				<institution content-type="orgname">Universidade Federal Fluminense</institution>
				<addr-line>
					<named-content content-type="city">Niterói</named-content>
					<named-content content-type="state">RJ</named-content>
				</addr-line>
				<country country="BR">Brazil</country>
				<institution content-type="original">Universidade Federal Fluminense, Niterói, RJ – Brazil</institution>
			</aff>
			<aff id="aff2">
				<label>2</label>
				<institution content-type="orgname">Faculdade de Ciências Médicas de Maricá</institution>
				<institution content-type="orgdiv1">Faculdade de Medicina</institution>
				<addr-line>
					<named-content content-type="city">Maricá</named-content>
					<named-content content-type="state">RJ</named-content>
				</addr-line>
				<country country="BR">Brazil</country>
				<institution content-type="original">Faculdade de Ciências Médicas de Maricá – Faculdade de Medicina, Maricá, RJ – Brazil</institution>
			</aff>
			<aff id="aff3">
				<label>3</label>
				<institution content-type="orgname">Universidade Federal Fluminense</institution>
				<institution content-type="orgdiv1">Hospital Universitário Antônio Pedro</institution>
				<addr-line>
					<named-content content-type="city">Niterói</named-content>
					<named-content content-type="state">RJ</named-content>
				</addr-line>
				<country country="BR">Brazil</country>
				<institution content-type="original">Universidade Federal Fluminense – Hospital Universitário Antônio Pedro, Niterói, RJ – Brazil</institution>
			</aff>
			<author-notes>
				<corresp id="c01">
					<label>Mailing Address:</label> Evandro Tinoco Mesquita • Universidade Federal Fluminense – R. Des. Athayde Parreiras, 100. Postal Code 24070-090, Fátima, Niterói, RJ – Brazil E-mail: <email>etmesquita@gmail.com</email>
				</corresp>
				<fn fn-type="coi-statement">
					<label>Potential Conflict of Interest:</label>
					<p>No potential conflict of interest relevant to this article was reported.</p>
				</fn>
				<fn fn-type="edited-by">
					<label>Editor responsible for the review:</label>
					<p>Victor Issa</p>
				</fn>
			</author-notes>
			<pub-date date-type="pub" publication-format="electronic">
				<day>21</day>
				<month>08</month>
				<year>2026</year>
			</pub-date>
			<pub-date date-type="collection" publication-format="electronic">
				<year>2026</year>
			</pub-date>
			<volume>6</volume>
			<elocation-id>e20260016</elocation-id>
			<history>
				<date date-type="received">
					<day>27</day>
					<month>05</month>
					<year>2026</year>
				</date>
				<date date-type="rev-recd">
					<day>16</day>
					<month>06</month>
					<year>2026</year>
				</date>
				<date date-type="accepted">
					<day>16</day>
					<month>06</month>
					<year>2026</year>
				</date>
			</history>
			<permissions>
				<license license-type="open-access" xlink:href="https://creativecommons.org/licenses/by/4.0/" xml:lang="en">
					<license-p> This is an Open Access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. </license-p>
				</license>
			</permissions>
			<abstract>
				<title>Abstract</title>
				<sec>
					<title>Background</title>
					<p> Myocarditis remains a diagnostic challenge, and interest in using artificial intelligence to support noninvasive diagnostic evaluation has grown rapidly. Although the literature at the intersection of artificial intelligence and myocarditis has expanded substantially, it remains fragmented and lacks comprehensive synthesis.</p>
				</sec>
				<sec>
					<title>Objectives</title>
					<p> To map the global scientific output on the application of artificial intelligence to myocarditis by identifying leading contributors, collaboration patterns, and emerging thematic trends.</p>
				</sec>
				<sec>
					<title>Methods</title>
					<p> We conducted a bibliometric analysis of articles on artificial intelligence applied to myocarditis indexed in the Web of Science Core Collection. The R package bibliometrix and VOSviewer were used to perform bibliometric network and thematic analyses.</p>
				</sec>
				<sec>
					<title>Results</title>
					<p> A total of 47 articles were included. Scientific output increased markedly after 2021, coinciding with the COVID-19 pandemic, and peaked in 2025. China led in publication volume, whereas the United Kingdom and Germany achieved the greatest citation impact. No publications originated from Latin America, the Caribbean, or Africa. International collaboration networks remain fragmented. Keyword and thematic analyses identified the integration of deep learning with cardiovascular magnetic resonance imaging as the principal motor theme, reflecting a shift away from traditional endomyocardial biopsy. The most highly cited authors were primarily computer and data scientists.</p>
				</sec>
				<sec>
					<title>Conclusions</title>
					<p> Research at the intersection of artificial intelligence and myocarditis is an emerging field undergoing methodological consolidation, with growth accelerated by the COVID-19 pandemic. Cardiovascular magnetic resonance imaging has become the primary data source for artificial intelligence-based applications. The principal challenge appears to be translational, underscoring the need for stronger collaboration between algorithm developers and clinicians. The successful integration of these models into routine clinical practice will likely depend on multidisciplinary collaboration and validation across diverse, multicenter populations.</p>
				</sec>
			</abstract>
			<kwd-group xml:lang="en">
				<title>Keywords</title>
				<kwd>Myocarditis</kwd>
				<kwd>Artificial Intelligence</kwd>
				<kwd>Machine Learning</kwd>
				<kwd>Deep Learning</kwd>
			</kwd-group>
			<funding-group>
				<funding-statement><bold>Sources of Funding:</bold>There were no external funding sources for this study</funding-statement>
			</funding-group>
			<counts>
				<fig-count count="16"/>
				<table-count count="2"/>
				<equation-count count="0"/>
				<ref-count count="29"/>
			</counts>
		</article-meta>
	</front>
	<body>
		<p>
					<fig id="f01">
						<label>Central Illustration:</label>
						<caption>
							<title>Global Mapping of Research on Artificial Intelligence Applied to Myocarditis</title>
						</caption>
						<graphic xlink:href="2764-3107-abchf-6-e20260016-gf01.tif"/>
						<attrib><italic>Descriptive bibliometric synthesis of 47 publications (2019-2026) indexed in the Web of Science Core Collection, illustrating annual scientific output, the thematic structure of the field (centered on CMR), geographic distribution, and the interdisciplinary authorship profile. AI: artificial intelligence; CMR: cardiovascular magnetic resonance; GAN: generative adversarial network; MCP: multiple-country publication.</italic></attrib>
					</fig>
				</p>
		<sec sec-type="intro">
			<title>Introduction</title>
			<p>Myocarditis is characterized by diffuse or focal inflammation of the myocardium, most commonly resulting from infectious agents, immune-mediated reactions, or chronic inflammatory processes.<sup><xref ref-type="bibr" rid="B1">1</xref></sup> Clinically, it remains a diagnostic challenge because of its heterogeneous and often nonspecific presentation, ranging from an asymptomatic course to heart failure, arrhythmias, and sudden death.<sup><xref ref-type="bibr" rid="B1">1</xref></sup> Traditionally, endomyocardial biopsy has been considered the diagnostic gold standard; however, advances in cardiovascular magnetic resonance imaging and the refinement of the Lake Louise criteria have transformed the noninvasive diagnostic approach.<sup><xref ref-type="bibr" rid="B1">1</xref></sup> Emerging clinical scenarios, particularly the expansion of cardio-oncology driven by the increasing incidence of myocarditis associated with immune checkpoint inhibitors, have further intensified the need for more accurate and timely diagnostic tools capable of supporting clinical decision-making while reducing reliance on invasive procedures.<sup><xref ref-type="bibr" rid="B2">2</xref></sup></p>
			<p>Artificial intelligence, including the subfields of machine learning and deep learning, has emerged as a promising tool for diagnostic support.<sup><xref ref-type="bibr" rid="B3">3</xref></sup> The ability of these algorithms to process large data sets, identify complex phenotypic patterns, and predict clinical outcomes suggests that artificial intelligence may advance precision medicine and contribute to the evolution of modern cardiology.<sup><xref ref-type="bibr" rid="B3">3</xref></sup> Despite growing technological interest, research at the intersection of artificial intelligence and myocarditis has expanded in a fragmented manner, with limited efforts to systematically characterize the field, identify leading research areas, map collaboration networks, or highlight emerging methodological trends.</p>
			<p>Bibliometric analysis provides a robust methodological framework for mapping the scientific landscape of a research field.<sup><xref ref-type="bibr" rid="B4">4</xref></sup> Unlike conventional systematic reviews, bibliometric approaches enable visualization of a field’s intellectual structure, identification of co-authorship networks, assessment of the geographic distribution of scientific production, and analysis of the temporal evolution of research topics.<sup><xref ref-type="bibr" rid="B4">4</xref>,<xref ref-type="bibr" rid="B5">5</xref></sup></p>
			<p>This study aimed to conduct a bibliometric analysis of the global scientific literature on the application of artificial intelligence to myocarditis. Specifically, we sought to identify the leading contributors (i.e., authors, countries, institutions, and journals) as well as the major thematic clusters within the field, thereby providing a comprehensive reference to guide future research and support the integration of artificial intelligence technologies into personalized cardiovascular medicine (<xref ref-type="fig" rid="f01">Central Illustration</xref>).</p>
		</sec>
		<sec sec-type="methods">
			<title>Methods</title>
			<sec>
				<title>Data source</title>
				<p>The Web of Science Core Collection was selected as the data source for this bibliometric analysis because of its extensive coverage of indexed scientific journals.<sup><xref ref-type="bibr" rid="B6">6</xref></sup> The following advanced search string was applied: (“myocarditis” OR “inflammatory cardiomyopathy”) AND (“artificial intelligence” OR “machine learning” OR “deep learning”). The search was performed on February 27, 2026. Only original research articles were included, with no restrictions on the year of publication.</p>
				<p>All eligible records were retrieved in a single search session, yielding 82 articles. Two independent review authors manually screened titles and abstracts to exclude duplicate records and studies not primarily focused on myocarditis, following standardized recommendations for bibliometric research in the medical literature.<sup><xref ref-type="bibr" rid="B7">7</xref></sup> This screening process resulted in a final data set of 47 articles for analysis. The bibliographic records were exported in plain text format and saved as download.txt files for subsequent processing.</p>
			</sec>
			<sec>
				<title>Bibliometric analysis</title>
				<p>Data processing, bibliometric analyses, and visualization were performed using R (version 4.5.2) and VOSviewer (version 1.6.20). The R environment, through the <italic>bibliometrix</italic> package, enabled comprehensive bibliometric analyses, including data extraction, network construction, and assessment of the scientific productivity and impact of countries, institutions, journals, authors, and co-cited references.<sup><xref ref-type="bibr" rid="B5">5</xref></sup> The results were presented using bibliometric network maps and graphical visualizations.</p>
				<p>Temporal analyses were also performed to examine the evolution of high-frequency keywords, providing insights into the development of research topics and emerging trends within the field. In addition, VOSviewer was used to construct and visualize co-citation and keyword co-occurrence networks.<sup><xref ref-type="bibr" rid="B8">8</xref></sup></p>
			</sec>
		</sec>
		<sec sec-type="results">
			<title>Results</title>
			<sec>
				<title>Publication analysis</title>
				<p>A total of 47 publications published between 2019 and 2026 were included in this study. The temporal distribution of publications shows modest scientific output through 2020, followed by sustained growth beginning in 2021. Annual publication volume increased from two articles in 2019 to 13 in 2025, representing the highest level of scientific production during the study period. The apparent decline observed in 2026 reflects the partial coverage of the year of publication, as the literature search was conducted on February 27, 2026.</p>
				<p>Citation patterns differed from publication trends. Scientific impact, measured by the mean number of citations per article per year, peaked in 2021 at 5.7 citations/year. Thereafter, the average annual citation rate progressively declined, reaching values close to zero for articles published in 2025 and 2026, reflecting the limited time available for newly published studies to accumulate citations (<xref ref-type="fig" rid="f02">Figure 1A</xref>).</p>
				<p>
					<fig id="f02">
						<label>Figure 1</label>
						<caption>
							<title>– Scientific production and geographic distribution. A) Annual scientific output and mean total citations per article. B) Scientific output by country, classified as SCPs and MCPs. MCP: multiple-country publication; SCP: single-country publication.</title>
						</caption>
						<graphic xlink:href="2764-3107-abchf-6-e20260016-gf02.tif"/>
					</fig>
				</p>
			</sec>
			<sec>
				<title>Country analysis</title>
				<p>Analysis of the corresponding authors’ affiliations showed that China was the leading contributor (n = 15 [31.9%]). Germany ranked second (n = 7 [14.9%]), followed by the United States (n = 6 [12.8%]) and the United Kingdom (n = 4 [8.5%]).</p>
				<p>Patterns of international collaboration varied considerably among the most productive countries. Germany exhibited the highest proportion of multiple-country publications (MCP), with 57.1% of its studies (four of seven) involving international collaboration. In contrast, China’s scientific output was predominantly domestic, comprising 11 single-country publications and an MCP rate of 26.7%. Among the leading publishing countries, the United States demonstrated the lowest level of international collaboration, with five of its six publications involving only domestic authors (MCP rate: 16.7%) (<xref ref-type="fig" rid="f02">Figure 1B</xref>).</p>
				<p>In terms of scientific impact, the United Kingdom ranked first, with 152 total citations and a mean of 38 citations per article. China ranked second in total citations (76), although its mean citation rate was substantially lower (5.1 citations/article). Germany accumulated 63 citations overall, corresponding to an average of nine citations per publication (<xref ref-type="table" rid="t1">Table 1</xref>).</p>
				<p>
					<table-wrap id="t1">
						<label>Table 1</label>
						<caption>
							<title>– Global leaders in scientific productivity in research on artificial intelligence applied to myocarditis</title>
						</caption>
						<table frame="hsides" rules="groups">
							<colgroup width="25%">
								<col/>
								<col/>
								<col/>
								<col/>
							</colgroup>
							<thead>
								<tr>
									<th>Rank</th>
									<th>Leading countries (total citations)</th>
									<th>Leading institutions (total articles)</th>
									<th>Most cited journals (total citations)</th>
								</tr>
							</thead>
							<tbody>
								<tr>
									<td>1</td>
									<td align="center">United Kingdom (152)</td>
									<td align="center">University of Bonn (11)</td>
									<td align="center">Circulation (58)</td>
								</tr>
								<tr>
									<td>2</td>
									<td align="center">China (76)</td>
									<td align="center">Taipei Medical University (10)</td>
									<td align="center">Journal of the American College of Cardiology (53)</td>
								</tr>
								<tr>
									<td>3</td>
									<td align="center">Germany (63)</td>
									<td align="center">University of London (8)</td>
									<td align="center">Journal of Cardiovascular Magnetic Resonance (42)</td>
								</tr>
								<tr>
									<td>4</td>
									<td align="center">Switzerland (36)</td>
									<td align="center">Mayo Clinic (7)</td>
									<td align="center">Radiology (42)</td>
								</tr>
								<tr>
									<td>5</td>
									<td align="center">United States (36)</td>
									<td align="center">Technical University of Munich (6)</td>
									<td align="center">European Heart Journal (34)</td>
								</tr>
								<tr>
									<td>6</td>
									<td align="center">Australia (32)</td>
									<td align="center">Imperial College London (5)</td>
									<td align="center">JACC: Cardiovascular Imaging (33)</td>
								</tr>
								<tr>
									<td>7</td>
									<td align="center">Italy (16)</td>
									<td align="center">Deakin University (4)</td>
									<td align="center">Lecture Notes in Computer Science (33)</td>
								</tr>
								<tr>
									<td>8</td>
									<td align="center">Austria (8)</td>
									<td align="center">Johannes Gutenberg University Mainz (4)</td>
									<td align="center">European Radiology (25)</td>
								</tr>
								<tr>
									<td>9</td>
									<td align="center">Iran (8)</td>
									<td align="center">Medical University of Innsbruck (4)</td>
									<td align="center">Scientific Reports (22)</td>
								</tr>
								<tr>
									<td>10</td>
									<td align="center">Korea (4)</td>
									<td align="center">Okayama University (4)</td>
									<td align="center">arXiv (21)</td>
								</tr>
							</tbody>
						</table>
						<table-wrap-foot>
							<fn id="TFN1">
								<p><italic>Note: Data represent the top 10 entities in each category based on publications indexed in the Web of Science Core Collection (2019-2026). Values in parentheses indicate total citations for countries and journals and the total number of original articles for institutions.</italic></p>
							</fn>
						</table-wrap-foot>
					</table-wrap>
				</p>
				<p>Despite producing fewer publications, Switzerland and Australia achieved high citation impact, with averages of 36 and 32 citations per article, respectively. The United States accumulated 36 total citations, corresponding to a mean of six citations per publication. Overall, scientific production was concentrated in China, Europe, and the United States.</p>
			</sec>
			<sec>
				<title>Affiliation analysis</title>
				<p>The institutions with the highest publication output are presented in <xref ref-type="table" rid="t1">Table 1</xref>. The University of Bonn ranked first with 11 publications, accounting for approximately 25% of the total sample, followed by Taipei Medical University (n = 10), University of London (n = 8), and Mayo Clinic (n = 7).</p>
				<p>Inter-institutional collaboration was characterized by compartmentalized research networks composed of clusters with limited interaction. The orange cluster represents a Euro-Australian collaboration network involving Deakin University and University of Granada, highlighting the contribution of research groups specializing in computer science and signal processing (<xref ref-type="fig" rid="f02">Figure 2A</xref>).</p>
				<p>The German network, represented by the red cluster, is characterized by dense regional collaboration among medical centers, led by Johannes Gutenberg University Mainz, University of Bonn, and Hannover Medical School. The green cluster represents the British research consortium, comprising strong collaborative links among the University of London, Imperial College London, and King’s College London.</p>
			</sec>
			<sec>
				<title>Author analysis</title>
				<p>Author productivity was led by Alizadehsani R (n = 4), followed by Górriz JM (n = 3). In terms of local impact, defined as the number of times a document within the analyzed data set was cited by other documents in the same data set, eight authors tied for the highest score, each with four local citations: Alizadehsani R, Alkadhi H, Di Noto T, Gantert E, Manka R, Mannil M, Soda P, and von Spiczak J.</p>
				<p>Application of Lotka’s Law demonstrated a marked deviation from the classical theoretical distribution.<sup><xref ref-type="bibr" rid="B9">9</xref></sup> Overall, 98.4% of authors contributed only a single publication, whereas only a small proportion authored two or more articles.</p>
				<p>The co-authorship network in artificial intelligence research on myocarditis was fragmented into eight distinct clusters, two of which were centered on authors with the highest local impact. The largest cluster (red) was organized around Alizadehsani R and Górriz JM, whereas the blue cluster comprised Kaya K, Luetkens JA, Pennig L, and Zimmer S. The network visualization highlights the presence of isolated collaborative groups with no external connections, with link thickness representing the frequency of direct co-authorship (<xref ref-type="fig" rid="f03">Figure 2B</xref>).</p>
				<p>
					<fig id="f03">
						<label>Figure 2</label>
						<caption>
							<title>– Author impact metrics.</title>
						</caption>
						<graphic xlink:href="2764-3107-abchf-6-e20260016-gf03.tif"/>
					</fig>
				</p>
				<p>Additional indicators of local scientific impact, including the H-index and M-index, further characterized author influence within the field. Alizadehsani R achieved the highest H-index (H = 4), with four publications receiving at least four citations each. Górriz JM ranked second (H = 3). After adjustment for academic age, Górriz JM attained the highest M-index (M = 1.0), followed by Alizadehsani R (M = 0.8). Three additional authors demonstrated comparable performance, each with an H-index of 2 and an M-index of 0.67: Kaya K, Luetkens JA, and Pennig L (<xref ref-type="fig" rid="f04">Figure 3</xref>).</p>
				<p>
					<fig id="f04">
						<label>Figure 3</label>
						<caption>
							<title>– Author impact metrics.</title>
						</caption>
						<graphic xlink:href="2764-3107-abchf-6-e20260016-gf04.tif"/>
					</fig>
				</p>
			</sec>
			<sec>
				<title>Journal analysis</title>
				<p>The distribution of scientific output by journal identified the Journal of Clinical Medicine as the most productive journal, with four published articles. However, analysis of local citation impact showed that Circulation ranked first in total citations (n = 58), followed by the Journal of the American College of Cardiology (n = 53), the Journal of Cardiovascular Magnetic Resonance (n = 42), and Radiology (n = 42) (<xref ref-type="table" rid="t1">Table 1</xref>).</p>
				<p>Application of Bradford’s Law characterized the dispersion of publications and identified the core journals in this research field.<sup><xref ref-type="bibr" rid="B10">10</xref></sup> The analysis revealed a small core (Zone 1) comprising four journals that concentrated the largest number of publications on the topic, each contributing two articles: i) CAAI Transactions on Intelligence Technology; ii) Diagnostics; iii) Journal of Cardiovascular Magnetic Resonance; and iv) Quantitative Imaging in Medicine and Surgery.</p>
			</sec>
			<sec>
				<title>Global citation analysis</title>
				<p>Global citation counts provide an indicator of an article’s scientific influence within the broader academic community. The most highly cited publication in the data set (n = 116) was the study first-authored by Joy G,<sup><xref ref-type="bibr" rid="B11">11</xref></sup> titled “Prospective Case-Control Study of Cardiovascular Abnormalities 6 Months Following Mild COVID-19 in Healthcare Workers,” published in the Journal of the American College of Cardiology. The study investigated cardiovascular abnormalities following SARS-CoV-2 infection using artificial intelligence-based analytical approaches.</p>
				<p>The second most highly cited article (n = 41), titled “CNN-KCL: Automatic myocarditis diagnosis using convolutional neural network combined with k-means clustering,” was first-authored by Sharifrazi D,<sup><xref ref-type="bibr" rid="B12">12</xref></sup> with Alizadehsani R as a co-author. The study proposed a deep learning framework for the automated diagnosis of myocarditis.</p>
			</sec>
			<sec>
				<title>Co-cited reference analysis</title>
				<p>The intellectual structure of research on artificial intelligence applied to myocarditis was mapped through co-citation network analysis at three levels: i) journals, ii) authors, and iii) individual references.</p>
				<p>The journal co-citation network exhibited a tripartite structure. The red cluster represents clinical cardiology and comprises high-impact journals, including Circulation, the Journal of the American College of Cardiology, and the European Heart Journal. The blue cluster corresponds to the cardiovascular imaging domain, encompassing specialized journals such as Radiology, the Journal of Cardiovascular Magnetic Resonance, and JACC: Cardiovascular Imaging. The green cluster reflects the technological and multidisciplinary dimension of the field, integrating journals from computer science and general scientific research, with Lecture Notes in Computer Science occupying a prominent position (<xref ref-type="fig" rid="f04">Figure 4A</xref>).</p>
				<p>The author co-citation network revealed two major intellectual fronts at the intersection of artificial intelligence and myocarditis. The blue cluster represents the clinical and diagnostic domain, centered on Ferreira VM, whose work on cardiovascular magnetic resonance has become a cornerstone in the diagnosis of myocardial inflammation (<xref ref-type="fig" rid="f04">Figure 4B</xref>).<sup><xref ref-type="bibr" rid="B13">13</xref></sup></p>
				<p>The red cluster represents the algorithmic and methodological domain, centered on authors whose work focuses on the development of computational models, including Moravvej SV,<sup><xref ref-type="bibr" rid="B14">14</xref></sup> Vakilian S,<sup><xref ref-type="bibr" rid="B15">15</xref></sup> and Hong L.<sup><xref ref-type="bibr" rid="B16">16</xref></sup> This cluster also includes prominent computer scientists, such as Cortes C, co-developer of support vector machines (SVMs).<sup><xref ref-type="bibr" rid="B17">17</xref></sup> The network demonstrates a strong structural connection between these two domains, reflecting the frequent co-citation of authors from both the medical and computational sciences.</p>
				<p>At the reference level, the co-citation network places the article by Ferreira VM (2018) at its center, linking multiple thematic clusters. This landmark publication introduced the updated Lake Louise criteria for the diagnosis of myocarditis using cardiovascular magnetic resonance imaging.<sup><xref ref-type="bibr" rid="B13">13</xref></sup> The network also highlights seminal methodological contributions, including the work of Cortes C,<sup><xref ref-type="bibr" rid="B17">17</xref></sup> which established the theoretical foundations of SVMs. Building upon these foundational studies, a dense and highly interconnected body of recent applied research has emerged, with notable contributions from Moravvej SV (2021, 2022), Danaei S (2022), Sudre CH (2017), and Hong L (2023) (<xref ref-type="fig" rid="f06">Figure 5</xref>).</p>
				<p>
					<fig id="f06">
						<label>Figure 5</label>
						<caption>
							<title>– Reference co-citation network.</title>
						</caption>
						<graphic xlink:href="2764-3107-abchf-6-e20260016-gf06.tif"/>
					</fig>
				</p>
			</sec>
			<sec>
				<title>Co-occurrence analysis</title>
				<p>The conceptual structure of the field was mapped through keyword co-occurrence analysis, identifying terms that frequently co-occurred across the articles included in the data set. Three high-frequency keywords (<italic>myocarditis, diagnosis</italic>, and <italic>artificial intelligence</italic>) served as the central nodes of the predominant thematic clusters (<xref ref-type="fig" rid="f07">Figure 6</xref>).</p>
				<p>
					<fig id="f07">
						<label>Figure 6</label>
						<caption>
							<title>– Reference co-citation network.</title>
						</caption>
						<graphic xlink:href="2764-3107-abchf-6-e20260016-gf07.tif"/>
					</fig>
				</p>
				<p>The red cluster is centered on <italic>myocarditis</italic> and links challenging clinical scenarios, including fulminant myocarditis and immune checkpoint inhibitor-associated myocarditis, with advanced computational approaches such as generative adversarial networks, reinforcement learning, and data augmentation techniques.</p>
				<p>The blue cluster is organized around <italic>diagnosis</italic> and comprises terms related to algorithmic development, including <italic>deep learning, classification</italic>, and <italic>prediction</italic>, which are strongly associated with cardiovascular magnetic resonance imaging. The green and purple clusters connect <italic>artificial intelligence</italic> and <italic>machine learning</italic> with differential diagnoses and imaging biomarkers, including cardiac sarcoidosis, viral myocarditis, and late gadolinium enhancement.</p>
				<p>The orange and brown clusters suggest the substantial influence of the COVID-19 pandemic on the evolution of the literature. The orange cluster is centered on <italic>COVID-19</italic> and <italic>SARS-CoV-2</italic>, whereas the brown and purple clusters reflect the influence of professional society recommendations, represented by terms such as <italic>position statement</italic> and <italic>working group</italic>, highlighting ongoing efforts to standardize the diagnosis of myocarditis. When the network was filtered to retain only the strongest associations, the most prominent and densely connected relationship was observed among the nodes <italic>diagnosis, cardiovascular magnetic resonance imaging</italic>, and <italic>myocarditis</italic>.</p>
			</sec>
			<sec>
				<title>Research hotspots and emerging trends</title>
				<p>The intellectual structure of the field and emerging research fronts were characterized using a thematic map, which classifies research domains into four quadrants according to density (degree of development) and centrality (relevance within the research network).</p>
				<p>The upper-right quadrant comprises the motor themes, characterized by both high centrality and high density, representing the most influential and well-developed topics in the field. The predominant themes include <italic>cardiovascular magnetic resonance, artificial intelligence, magnetic resonance imaging, outcomes</italic>, and <italic>position statement</italic> (<xref ref-type="fig" rid="f08">Figure 7</xref>).</p>
				<p>
					<fig id="f08">
						<label>Figure 7</label>
						<caption>
							<title>– Thematic map of research trends and emerging hotspots.</title>
						</caption>
						<graphic xlink:href="2764-3107-abchf-6-e20260016-gf08.tif"/>
					</fig>
				</p>
				<p>The lower-right quadrant contains the basic themes, which are highly relevant to the field but exhibit a lower degree of internal development. This quadrant includes <italic>deep learning, magnetic resonance imaging</italic>, and <italic>strain</italic>. At the interface with the motor themes are clusters related to <italic>myocarditis, COVID-19, SARS-CoV-2, disease</italic>, and <italic>fulminant myocarditis</italic>.</p>
				<p>
					<fig id="f05">
						<label>Figure 4</label>
						<caption>
							<title>– Journal and author co-citation networks. A) Journal co-citation network. B) Author co-citation network.</title>
						</caption>
						<graphic xlink:href="2764-3107-abchf-6-e20260016-gf05.tif"/>
					</fig>
				</p>
				<p>The upper-left quadrant contains the niche themes, which are highly developed but remain peripheral to the overall research network. These include <italic>endomyocardial biopsy</italic> and <italic>inflammatory cardiomyopathy</italic>. The lower-left quadrant comprises emerging or declining themes, whose interpretation depends on their temporal evolution. This region is dominated by broad prediction-oriented terms, including <italic>classification, prediction, diagnosis</italic>, and <italic>machine learning</italic>.</p>
				<p>Temporal analyses based on keyword frequency and trend-topic plots identified a clear inflection point in 2021, corresponding to the marked increase in scientific output and driven by the rapid growth of the terms <italic>myocarditis, diagnosis</italic>, and <italic>artificial intelligence</italic>. From 2023 onward, <italic>deep learning</italic> and <italic>cardiovascular magnetic resonance</italic> became increasingly prominent. Among all keywords, <italic>machine learning</italic> exhibited the longest period of sustained use throughout the study interval.</p>
			</sec>
		</sec>
		<sec sec-type="discussion">
			<title>Discussion</title>
			<p>This bibliometric analysis indicates that the intersection of artificial intelligence and myocarditis represents an emerging and still relatively immature research field. Publication output increased markedly after 2021, coinciding with the COVID-19 pandemic, which likely served as a global catalyst for research in this area. The heightened clinical interest in virus-induced myocardial injury, together with the availability of large hospital data sets generated during the pandemic, may have accelerated the adoption of machine learning and deep learning algorithms for risk stratification and diagnostic support in patients with suspected myocarditis.<sup><xref ref-type="bibr" rid="B11">11</xref>,<xref ref-type="bibr" rid="B18">18</xref></sup></p>
			<p>The recent decline in the mean number of citations per article should not be interpreted as evidence of diminishing scientific relevance. A more plausible explanation is citation latency, an inherent limitation of bibliometric analyses.<sup><xref ref-type="bibr" rid="B19">19</xref></sup> Articles published within the past 24 months have had limited time to accumulate citations. Accordingly, the citation peak observed in 2021 likely reflects the methodological foundation upon which subsequent research in the field has been built.</p>
			<p>Nevertheless, the current growth rate remains modest compared with the broader field of cardiology.<sup><xref ref-type="bibr" rid="B20">20</xref></sup> This finding suggests that artificial intelligence applications in myocarditis are still undergoing methodological consolidation. Continued expansion of artificial intelligence in cardiology is anticipated over the coming decades, particularly in cardiovascular imaging, where artificial intelligence-based approaches are expected to play an increasingly prominent role.<sup><xref ref-type="bibr" rid="B3">3</xref>,<xref ref-type="bibr" rid="B21">21</xref></sup></p>
			<p>The analysis of country-level contributions revealed distinct research strategies among the leading publishing nations. China accounted for the largest number of publications, suggesting a strong national emphasis on the development and implementation of artificial intelligence-based diagnostic technologies. In contrast, the European research network, led by the United Kingdom and Germany, produced fewer publications but achieved substantially greater citation impact, with most of the highly cited studies originating from these countries. The United States exhibited a predominantly domestic research profile, possibly reflecting greater reliance on institutional and private data sets than on international collaborative consortia. A notable finding was the complete absence of publications from Latin America, the Caribbean, and Africa. This geographic imbalance raises important concerns regarding the external validity and generalizability of artificial intelligence models for myocarditis, as training and validation data sets may inadequately represent these populations, thereby increasing the risk of algorithmic bias in clinical decision-support systems.<sup><xref ref-type="bibr" rid="B22">22</xref></sup></p>
			<p>The authorship structure of the field also remains highly fragmented. According to Lotka’s Law, 98.4% of authors contributed only a single publication on the topic, indicating a research landscape characterized by isolated contributions rather than sustained activity from dedicated research groups.<sup><xref ref-type="bibr" rid="B9">9</xref></sup> Examination of the academic backgrounds of the authors with the greatest local impact highlights another defining characteristic of the field: its pronounced interdisciplinarity and the leading role currently played by researchers from engineering, computer science, and other quantitative disciplines in advancing medical artificial intelligence.<sup><xref ref-type="bibr" rid="B23">23</xref></sup> Juan Manuel Górriz, who achieved the highest M-index (1.0), has a background in Physics and Electronic Engineering and specializes in statistical signal processing for biomedical applications.<sup><xref ref-type="bibr" rid="B24">24</xref></sup> Roohallah Alizadehsani, who led the field in both publication count and H-index, holds a doctorate in Computer Engineering and focuses on the development of machine learning algorithms for cardiovascular diseases.<sup><xref ref-type="bibr" rid="B25">25</xref></sup></p>
			<p>This profile suggests that the methodological and technological development of artificial intelligence models is currently outpacing their clinical validation. The prominent role of engineers and data scientists indicates that research has primarily focused on the design, training, and optimization of computational models capable of processing complex data sets, particularly cardiovascular imaging and other high-dimensional clinical data.<sup><xref ref-type="bibr" rid="B26">26</xref></sup></p>
			<p>A similar pattern is evident in the leading journals within the field. Scientific dissemination is concentrated in journals focused on innovation in cardiovascular imaging and diagnostic technologies, reflecting the reliance of current artificial intelligence models on imaging-derived biomarkers for myocarditis detection. The translation of these technologies into routine clinical practice, however, will likely require a greater emphasis on translational research.<sup><xref ref-type="bibr" rid="B3">3</xref></sup> Bridging the gap between computational development and bedside implementation will depend on multidisciplinary teams in which expertise in algorithm development is integrated with the pathophysiological interpretation and rigorous data curation provided by cardiologists and clinical investigators.</p>
			<p>This evolution in the diagnostic paradigm is further reflected in the thematic map and keyword co-occurrence networks. Although endomyocardial biopsy remains the traditional diagnostic gold standard, it is positioned within a niche thematic cluster, whereas cardiovascular imaging techniques dominate the motor themes. Current research is increasingly focused on refining deep learning and machine learning algorithms to identify subtle imaging patterns and improve the performance of noninvasive diagnostic approaches.<sup><xref ref-type="bibr" rid="B3">3</xref>,<xref ref-type="bibr" rid="B13">13</xref>,<xref ref-type="bibr" rid="B21">21</xref></sup></p>
			<p>Beyond improving diagnostic accuracy, recent studies have explored the use of generative artificial intelligence and data augmentation techniques to address the limited availability of data for rare or underdiagnosed clinical presentations, including fulminant myocarditis and immune checkpoint inhibitor-associated myocarditis.<sup><xref ref-type="bibr" rid="B27">27</xref>,<xref ref-type="bibr" rid="B28">28</xref></sup> In parallel, the prominence of the keyword <italic>outcomes</italic> among the motor themes reflects growing interest in prognostic modeling and clinical event prediction. Finally, the co-occurrence of artificial intelligence-related terms with guideline-oriented keywords, such as <italic>position statement</italic> and <italic>working group</italic>, suggests an increasing emphasis on establishing standardized frameworks for the evaluation, validation, and reporting of artificial intelligence applications in cardiovascular imaging.<sup><xref ref-type="bibr" rid="B29">29</xref></sup></p>
			<sec>
				<title>Limitations</title>
				<p>Several limitations of this study should be acknowledged. First, data extraction was restricted to the Web of Science Core Collection; therefore, relevant articles indexed exclusively in databases such as Scopus, Embase, or PubMed may not have been captured. Nevertheless, the Web of Science was selected because it provides the standardized metadata required for robust bibliometric and citation network analyses.<sup><xref ref-type="bibr" rid="B6">6</xref></sup></p>
				<p>Second, the search strategy relied on predefined keywords. Consequently, studies describing specific computational techniques (e.g., random forests or SVMs) without explicitly including the broader terms <italic>artificial intelligence</italic> or <italic>machine learning</italic> in the title, abstract, or keywords may have been excluded.</p>
				<p>Third, bibliometric analyses are inherently subject to citation latency bias,<sup><xref ref-type="bibr" rid="B19">19</xref></sup> whereby recently published articles have had insufficient time to accumulate citations regardless of their scientific quality or potential impact. This limitation was partially addressed by incorporating the M-index, which accounts for the time-dependent nature of citation accumulation.</p>
				<p>Finally, the relatively small number of included publications (n = 47) may limit the generalizability of the findings. However, this characteristic also reflects the early developmental stage of artificial intelligence research applied to myocarditis, highlighting the emergence of a specialized and still-expanding research field.</p>
			</sec>
		</sec>
		<sec sec-type="conclusions">
			<title>Conclusion</title>
			<p>This bibliometric analysis provides a comprehensive overview of the global scientific landscape at the intersection of artificial intelligence and myocarditis. The field remains in an early stage of development, is characterized by marked geographic disparities, and is currently driven predominantly by researchers from engineering and computer science rather than by clinical cardiologists. Its recent expansion coincided with the COVID-19 pandemic, while cardiovascular magnetic resonance imaging has emerged as the primary data source underpinning artificial intelligence-based diagnostic models.</p>
			<p>The successful translation of these technologies into routine clinical practice will likely depend on multidisciplinary collaboration and rigorous validation across diverse populations, particularly in regions that remain underrepresented in the current literature, such as Latin America, the Caribbean, and Africa. Although this study was limited to a single bibliographic database and included a relatively small number of publications, these findings provide a valuable baseline for future investigations and contribute to a better understanding of the evolution, current status, and future directions of this rapidly developing field.</p>
		</sec>
	</body>
	<back>
		<ref-list>
			<title>References</title>
			<ref id="B1">
				<label>1</label>
				<mixed-citation>1. Tschöpe C, Ammirati E, Bozkurt B, Caforio ALP, Cooper LT, Felix SB, et al. Myocarditis and Inflammatory Cardiomyopathy: Current Evidence and Future Directions. Nat Rev Cardiol. 2021;18(3):169-93. doi: 10.1038/s41569-020-00435-x.</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Tschöpe</surname>
							<given-names>C</given-names>
						</name>
						<name>
							<surname>Ammirati</surname>
							<given-names>E</given-names>
						</name>
						<name>
							<surname>Bozkurt</surname>
							<given-names>B</given-names>
						</name>
						<name>
							<surname>Caforio</surname>
							<given-names>ALP</given-names>
						</name>
						<name>
							<surname>Cooper</surname>
							<given-names>LT</given-names>
						</name>
						<name>
							<surname>Felix</surname>
							<given-names>SB</given-names>
						</name>
						<etal>et al</etal>
					</person-group>
					<article-title>Myocarditis and Inflammatory Cardiomyopathy: Current Evidence and Future Directions</article-title>
					<source>Nat Rev Cardiol</source>
					<year>2021</year>
					<volume>18</volume>
					<issue>3</issue>
					<fpage>169</fpage>
					<lpage>193</lpage>
					<pub-id pub-id-type="doi">10.1038/s41569-020-00435-x</pub-id>
				</element-citation>
			</ref>
			<ref id="B2">
				<label>2</label>
				<mixed-citation>2. Jensen G, Wang X, Kuempel J, Palaskas N, Chen Z, Yu W, et al. Immune Checkpoint Inhibitor-Associated Myocarditis: A Historical and Comprehensive Review. Am J Physiol Heart Circ Physiol. 2025;328(4):H734-51. doi: 10.1152/ajpheart.00687.2024.</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Jensen</surname>
							<given-names>G</given-names>
						</name>
						<name>
							<surname>Wang</surname>
							<given-names>X</given-names>
						</name>
						<name>
							<surname>Kuempel</surname>
							<given-names>J</given-names>
						</name>
						<name>
							<surname>Palaskas</surname>
							<given-names>N</given-names>
						</name>
						<name>
							<surname>Chen</surname>
							<given-names>Z</given-names>
						</name>
						<name>
							<surname>Yu</surname>
							<given-names>W</given-names>
						</name>
						<etal>et al</etal>
					</person-group>
					<article-title>Immune Checkpoint Inhibitor-Associated Myocarditis: A Historical and Comprehensive Review</article-title>
					<source>Am J Physiol Heart Circ Physiol</source>
					<year>2025</year>
					<volume>328</volume>
					<issue>4</issue>
					<fpage>H734</fpage>
					<lpage>H751</lpage>
					<pub-id pub-id-type="doi">10.1152/ajpheart.00687.2024</pub-id>
				</element-citation>
			</ref>
			<ref id="B3">
				<label>3</label>
				<mixed-citation>3. Rajpurkar P, Chen E, Banerjee O, Topol EJ. AI in Health and Medicine. Nat Med. 2022;28(1):31-8. doi: 10.1038/s41591-021-01614-0.</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Rajpurkar</surname>
							<given-names>P</given-names>
						</name>
						<name>
							<surname>Chen</surname>
							<given-names>E</given-names>
						</name>
						<name>
							<surname>Banerjee</surname>
							<given-names>O</given-names>
						</name>
						<name>
							<surname>Topol</surname>
							<given-names>EJ</given-names>
						</name>
					</person-group>
					<article-title>AI in Health and Medicine</article-title>
					<source>Nat Med</source>
					<year>2022</year>
					<volume>28</volume>
					<issue>1</issue>
					<fpage>31</fpage>
					<lpage>38</lpage>
					<pub-id pub-id-type="doi">10.1038/s41591-021-01614-0</pub-id>
				</element-citation>
			</ref>
			<ref id="B4">
				<label>4</label>
				<mixed-citation>4. Donthu N, Kumar S, Mukherjee D, Pandey N, Lim WM. How to Conduct a Bibliometric Analysis: An Overview and Guidelines. J Bus Res. 2021;133:285-96. doi: 10.1016/j.jbusres.2021.04.070.</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Donthu</surname>
							<given-names>N</given-names>
						</name>
						<name>
							<surname>Kumar</surname>
							<given-names>S</given-names>
						</name>
						<name>
							<surname>Mukherjee</surname>
							<given-names>D</given-names>
						</name>
						<name>
							<surname>Pandey</surname>
							<given-names>N</given-names>
						</name>
						<name>
							<surname>Lim</surname>
							<given-names>WM</given-names>
						</name>
					</person-group>
					<article-title>How to Conduct a Bibliometric Analysis: An Overview and Guidelines</article-title>
					<source>J Bus Res</source>
					<year>2021</year>
					<volume>133</volume>
					<fpage>285</fpage>
					<lpage>296</lpage>
					<pub-id pub-id-type="doi">10.1016/j.jbusres.2021.04.070</pub-id>
				</element-citation>
			</ref>
			<ref id="B5">
				<label>5</label>
				<mixed-citation>5. Aria M, Cuccurullo C. Bibliometrix: An R-Tool for Comprehensive Science Mapping Analysis. J Informetr. 2017;11(4):959-75. doi: 10.1016/j.joi.2017.08.007.</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Aria</surname>
							<given-names>M</given-names>
						</name>
						<name>
							<surname>Cuccurullo</surname>
							<given-names>C</given-names>
						</name>
					</person-group>
					<article-title>Bibliometrix: An R-Tool for Comprehensive Science Mapping Analysis</article-title>
					<source>J Informetr</source>
					<year>2017</year>
					<volume>11</volume>
					<issue>4</issue>
					<fpage>959</fpage>
					<lpage>975</lpage>
					<pub-id pub-id-type="doi">10.1016/j.joi.2017.08.007</pub-id>
				</element-citation>
			</ref>
			<ref id="B6">
				<label>6</label>
				<mixed-citation>6. Birkle C, Pendlebury DA, Schnell J, Adams J. Web of Science as a Data Source for Research on Scientific and Scholarly Activity. Quant Sci Stud. 2020;1(1):363-76. doi: 10.1162/qss_a_00018.</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Birkle</surname>
							<given-names>C</given-names>
						</name>
						<name>
							<surname>Pendlebury</surname>
							<given-names>DA</given-names>
						</name>
						<name>
							<surname>Schnell</surname>
							<given-names>J</given-names>
						</name>
						<name>
							<surname>Adams</surname>
							<given-names>J</given-names>
						</name>
					</person-group>
					<article-title>Web of Science as a Data Source for Research on Scientific and Scholarly Activity</article-title>
					<source>Quant Sci Stud</source>
					<year>2020</year>
					<volume>1</volume>
					<issue>1</issue>
					<fpage>363</fpage>
					<lpage>376</lpage>
					<pub-id pub-id-type="doi">10.1162/qss_a_00018</pub-id>
				</element-citation>
			</ref>
			<ref id="B7">
				<label>7</label>
				<mixed-citation>7. Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 Statement: An Updated Guideline for Reporting Systematic Reviews. BMJ. 2021;372:n71. doi: 10.1136/bmj.n71.</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Page</surname>
							<given-names>MJ</given-names>
						</name>
						<name>
							<surname>McKenzie</surname>
							<given-names>JE</given-names>
						</name>
						<name>
							<surname>Bossuyt</surname>
							<given-names>PM</given-names>
						</name>
						<name>
							<surname>Boutron</surname>
							<given-names>I</given-names>
						</name>
						<name>
							<surname>Hoffmann</surname>
							<given-names>TC</given-names>
						</name>
						<name>
							<surname>Mulrow</surname>
							<given-names>CD</given-names>
						</name>
						<etal>et al</etal>
					</person-group>
					<article-title>The PRISMA 2020 Statement: An Updated Guideline for Reporting Systematic Reviews</article-title>
					<source>BMJ</source>
					<year>2021</year>
					<volume>372</volume>
					<size units="pages">n71</size>
					<pub-id pub-id-type="doi">10.1136/bmj.n71</pub-id>
				</element-citation>
			</ref>
			<ref id="B8">
				<label>8</label>
				<mixed-citation>8. van Eck NJ, Waltman L. Software Survey: VOSviewer, a Computer Program for Bibliometric Mapping. Scientometrics. 2010;84(2):523-38. doi: 10.1007/s11192-009-0146-3.</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>van Eck</surname>
							<given-names>NJ</given-names>
						</name>
						<name>
							<surname>Waltman</surname>
							<given-names>L</given-names>
						</name>
					</person-group>
					<article-title>Software Survey: VOSviewer, a Computer Program for Bibliometric Mapping</article-title>
					<source>Scientometrics</source>
					<year>2010</year>
					<volume>84</volume>
					<issue>2</issue>
					<fpage>523</fpage>
					<lpage>538</lpage>
					<pub-id pub-id-type="doi">10.1007/s11192-009-0146-3</pub-id>
				</element-citation>
			</ref>
			<ref id="B9">
				<label>9</label>
				<mixed-citation>9. Lotka AJ. The Frequency Distribution of Scientific Productivity. J Wash Acad Sci. 1926;16:317-23.</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Lotka</surname>
							<given-names>AJ</given-names>
						</name>
					</person-group>
					<article-title>The Frequency Distribution of Scientific Productivity</article-title>
					<source>J Wash Acad Sci</source>
					<year>1926</year>
					<volume>16</volume>
					<fpage>317</fpage>
					<lpage>323</lpage>
				</element-citation>
			</ref>
			<ref id="B10">
				<label>10</label>
				<mixed-citation>10. Bradford SC. Sources of Information on Specific Subjects. Engineering. 1934;137:85-6.</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Bradford</surname>
							<given-names>SC</given-names>
						</name>
					</person-group>
					<article-title>Sources of Information on Specific Subjects</article-title>
					<source>Engineering</source>
					<year>1934</year>
					<volume>137</volume>
					<fpage>85</fpage>
					<lpage>86</lpage>
				</element-citation>
			</ref>
			<ref id="B11">
				<label>11</label>
				<mixed-citation>11. Joy G, Artico J, Kurdi H, Seraphim A, Lau C, Thornton GD, et al. Prospective Case-Control Study of Cardiovascular Abnormalities 6 Months Following Mild COVID-19 in Healthcare Workers. JACC Cardiovasc Imaging. 2021;14(11):2155-66. doi: 10.1016/j.jcmg.2021.04.011.</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Joy</surname>
							<given-names>G</given-names>
						</name>
						<name>
							<surname>Artico</surname>
							<given-names>J</given-names>
						</name>
						<name>
							<surname>Kurdi</surname>
							<given-names>H</given-names>
						</name>
						<name>
							<surname>Seraphim</surname>
							<given-names>A</given-names>
						</name>
						<name>
							<surname>Lau</surname>
							<given-names>C</given-names>
						</name>
						<name>
							<surname>Thornton</surname>
							<given-names>GD</given-names>
						</name>
						<etal>et al</etal>
					</person-group>
					<article-title>Prospective Case-Control Study of Cardiovascular Abnormalities 6 Months Following Mild COVID-19 in Healthcare Workers</article-title>
					<source>JACC Cardiovasc Imaging</source>
					<year>2021</year>
					<volume>14</volume>
					<issue>11</issue>
					<fpage>2155</fpage>
					<lpage>2166</lpage>
					<pub-id pub-id-type="doi">10.1016/j.jcmg.2021.04.011</pub-id>
				</element-citation>
			</ref>
			<ref id="B12">
				<label>12</label>
				<mixed-citation>12. Sharifrazi D, Alizadehsani R, Joloudari JH, Band SS, Hussain S, Sani ZA, et al. CNN-KCL: Automatic Myocarditis Diagnosis Using Convolutional Neural Network Combined with k-Means Clustering. Math Biosci Eng. 2022;19(3):2381-402. doi: 10.3934/mbe.2022110.</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Sharifrazi</surname>
							<given-names>D</given-names>
						</name>
						<name>
							<surname>Alizadehsani</surname>
							<given-names>R</given-names>
						</name>
						<name>
							<surname>Joloudari</surname>
							<given-names>JH</given-names>
						</name>
						<name>
							<surname>Band</surname>
							<given-names>SS</given-names>
						</name>
						<name>
							<surname>Hussain</surname>
							<given-names>S</given-names>
						</name>
						<name>
							<surname>Sani</surname>
							<given-names>ZA</given-names>
						</name>
						<etal>et al</etal>
					</person-group>
					<article-title>CNN-KCL: Automatic Myocarditis Diagnosis Using Convolutional Neural Network Combined with k-Means Clustering</article-title>
					<source>Math Biosci Eng</source>
					<year>2022</year>
					<volume>19</volume>
					<issue>3</issue>
					<fpage>2381</fpage>
					<lpage>2402</lpage>
					<pub-id pub-id-type="doi">10.3934/mbe.2022110</pub-id>
				</element-citation>
			</ref>
			<ref id="B13">
				<label>13</label>
				<mixed-citation>13. Ferreira VM, Schulz-Menger J, Holmvang G, Kramer CM, Carbone I, Sechtem U, et al. Cardiovascular Magnetic Resonance in Nonischemic Myocardial Inflammation: Expert Recommendations. J Am Coll Cardiol. 2018;72(24):3158-76. doi: 10.1016/j.jacc.2018.09.072.</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Ferreira</surname>
							<given-names>VM</given-names>
						</name>
						<name>
							<surname>Schulz-Menger</surname>
							<given-names>J</given-names>
						</name>
						<name>
							<surname>Holmvang</surname>
							<given-names>G</given-names>
						</name>
						<name>
							<surname>Kramer</surname>
							<given-names>CM</given-names>
						</name>
						<name>
							<surname>Carbone</surname>
							<given-names>I</given-names>
						</name>
						<name>
							<surname>Sechtem</surname>
							<given-names>U</given-names>
						</name>
						<etal>et al</etal>
					</person-group>
					<article-title>Cardiovascular Magnetic Resonance in Nonischemic Myocardial Inflammation: Expert Recommendations</article-title>
					<source>J Am Coll Cardiol</source>
					<year>2018</year>
					<volume>72</volume>
					<issue>24</issue>
					<fpage>3158</fpage>
					<lpage>3176</lpage>
					<pub-id pub-id-type="doi">10.1016/j.jacc.2018.09.072</pub-id>
				</element-citation>
			</ref>
			<ref id="B14">
				<label>14</label>
				<mixed-citation>14. Moravvej SV, Alizadehsani R, Khanam S, Sobhaninia Z, Shoeibi A, Khozeimeh F, et al. RLMD-PA: A Reinforcement Learning-Based Myocarditis Diagnosis Combined with a Population-Based Algorithm for Pretraining Weights. Contrast Media Mol Imaging. 2022;2022:8733632. doi: 10.1155/2022/8733632.</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Moravvej</surname>
							<given-names>SV</given-names>
						</name>
						<name>
							<surname>Alizadehsani</surname>
							<given-names>R</given-names>
						</name>
						<name>
							<surname>Khanam</surname>
							<given-names>S</given-names>
						</name>
						<name>
							<surname>Sobhaninia</surname>
							<given-names>Z</given-names>
						</name>
						<name>
							<surname>Shoeibi</surname>
							<given-names>A</given-names>
						</name>
						<name>
							<surname>Khozeimeh</surname>
							<given-names>F</given-names>
						</name>
						<etal>et al</etal>
					</person-group>
					<article-title>RLMD-PA: A Reinforcement Learning-Based Myocarditis Diagnosis Combined with a Population-Based Algorithm for Pretraining Weights</article-title>
					<source>Contrast Media Mol Imaging</source>
					<year>2022</year>
					<volume>2022</volume>
					<size units="pages">8733632</size>
					<pub-id pub-id-type="doi">10.1155/2022/8733632</pub-id>
				</element-citation>
			</ref>
			<ref id="B15">
				<label>15</label>
				<mixed-citation>15. Vakilian S, Moravvej SV, Fanian A. Using the Cuckoo Algorithm to Optimizing the Response Time and Energy Consumption Cost of Fog Nodes by Considering Collaboration in the Fog Layer. Int Conf Internet Things Appl. 2021:1-5. doi:10.1109/IoT52625.2021.9469722.</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Vakilian</surname>
							<given-names>S</given-names>
						</name>
						<name>
							<surname>Moravvej</surname>
							<given-names>SV</given-names>
						</name>
						<name>
							<surname>Fanian</surname>
							<given-names>A</given-names>
						</name>
					</person-group>
					<article-title>Using the Cuckoo Algorithm to Optimizing the Response Time and Energy Consumption Cost of Fog Nodes by Considering Collaboration in the Fog Layer</article-title>
					<source>Int Conf Internet Things Appl</source>
					<year>2021</year>
					<fpage>1</fpage>
					<lpage>5</lpage>
					<pub-id pub-id-type="doi">10.1109/IoT52625.2021.9469722</pub-id>
				</element-citation>
			</ref>
			<ref id="B16">
				<label>16</label>
				<mixed-citation>16. Hong L, Modirrousta MH, Nasirpour MH, Chargari MM, Mohammadi F, Moravvej SV, et al. GAN-LSTM-3D: An Efficient Method for Lung Tumour 3D Reconstruction Enhanced by Attention-Based LSTM. CAAI Trans Intell Technol. 2023:1-10. doi: 10.1049/cit2.12223.</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Hong</surname>
							<given-names>L</given-names>
						</name>
						<name>
							<surname>Modirrousta</surname>
							<given-names>MH</given-names>
						</name>
						<name>
							<surname>Nasirpour</surname>
							<given-names>MH</given-names>
						</name>
						<name>
							<surname>Chargari</surname>
							<given-names>MM</given-names>
						</name>
						<name>
							<surname>Mohammadi</surname>
							<given-names>F</given-names>
						</name>
						<name>
							<surname>Moravvej</surname>
							<given-names>SV</given-names>
						</name>
						<etal>et al</etal>
					</person-group>
					<article-title>GAN-LSTM-3D: An Efficient Method for Lung Tumour 3D Reconstruction Enhanced by Attention-Based LSTM</article-title>
					<source>CAAI Trans Intell Technol</source>
					<year>2023</year>
					<fpage>1</fpage>
					<lpage>10</lpage>
					<pub-id pub-id-type="doi">10.1049/cit2.12223</pub-id>
				</element-citation>
			</ref>
			<ref id="B17">
				<label>17</label>
				<mixed-citation>17. Cortes C, Vapnik V. Support-Vector Networks. Mach Learn. 1995;20:273-97. doi: 10.1007/BF00994018.</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Cortes</surname>
							<given-names>C</given-names>
						</name>
						<name>
							<surname>Vapnik</surname>
							<given-names>V</given-names>
						</name>
					</person-group>
					<article-title>Support-Vector Networks</article-title>
					<source>Mach Learn</source>
					<year>1995</year>
					<volume>20</volume>
					<fpage>273</fpage>
					<lpage>297</lpage>
					<pub-id pub-id-type="doi">10.1007/BF00994018</pub-id>
				</element-citation>
			</ref>
			<ref id="B18">
				<label>18</label>
				<mixed-citation>18. Krittanawong C, Virk HUH, Bangalore S, Wang Z, Johnson KW, Pinotti R, et al. Machine Learning Prediction in Cardiovascular Diseases: A Meta-Analysis. Sci Rep. 2020;10(1):16057. doi: 10.1038/s41598-020-72685-1.</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Krittanawong</surname>
							<given-names>C</given-names>
						</name>
						<name>
							<surname>Virk</surname>
							<given-names>HUH</given-names>
						</name>
						<name>
							<surname>Bangalore</surname>
							<given-names>S</given-names>
						</name>
						<name>
							<surname>Wang</surname>
							<given-names>Z</given-names>
						</name>
						<name>
							<surname>Johnson</surname>
							<given-names>KW</given-names>
						</name>
						<name>
							<surname>Pinotti</surname>
							<given-names>R</given-names>
						</name>
						<etal>et al</etal>
					</person-group>
					<article-title>Machine Learning Prediction in Cardiovascular Diseases: A Meta-Analysis</article-title>
					<source>Sci Rep</source>
					<year>2020</year>
					<volume>10</volume>
					<issue>1</issue>
					<size units="pages">16057</size>
					<pub-id pub-id-type="doi">10.1038/s41598-020-72685-1</pub-id>
				</element-citation>
			</ref>
			<ref id="B19">
				<label>19</label>
				<mixed-citation>19. Wang D, Song C, Barabási AL. Quantifying Long-Term Scientific Impact. Science. 2013;342(6154):127-32. doi: 10.1126/science.1237825.</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Wang</surname>
							<given-names>D</given-names>
						</name>
						<name>
							<surname>Song</surname>
							<given-names>C</given-names>
						</name>
						<name>
							<surname>Barabási</surname>
							<given-names>AL</given-names>
						</name>
					</person-group>
					<article-title>Quantifying Long-Term Scientific Impact</article-title>
					<source>Science</source>
					<year>2013</year>
					<volume>342</volume>
					<issue>6154</issue>
					<fpage>127</fpage>
					<lpage>132</lpage>
					<pub-id pub-id-type="doi">10.1126/science.1237825</pub-id>
				</element-citation>
			</ref>
			<ref id="B20">
				<label>20</label>
				<mixed-citation>20. Zhang Y, Xie J, Fu E, Cai W, Xu W. Artificial Intelligence in Cardiology: A Bibliometric Study. Am J Transl Res. 2024;16(3):1029-35. doi: 10.62347/HSFE6936.</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Zhang</surname>
							<given-names>Y</given-names>
						</name>
						<name>
							<surname>Xie</surname>
							<given-names>J</given-names>
						</name>
						<name>
							<surname>Fu</surname>
							<given-names>E</given-names>
						</name>
						<name>
							<surname>Cai</surname>
							<given-names>W</given-names>
						</name>
						<name>
							<surname>Xu</surname>
							<given-names>W</given-names>
						</name>
					</person-group>
					<article-title>Artificial Intelligence in Cardiology: A Bibliometric Study</article-title>
					<source>Am J Transl Res</source>
					<year>2024</year>
					<volume>16</volume>
					<issue>3</issue>
					<fpage>1029</fpage>
					<lpage>1035</lpage>
					<pub-id pub-id-type="doi">10.62347/HSFE6936</pub-id>
				</element-citation>
			</ref>
			<ref id="B21">
				<label>21</label>
				<mixed-citation>21. Lim LJ, Tison GH, Delling FN. Artificial Intelligence in Cardiovascular Imaging. Methodist Debakey Cardiovasc J. 2020;16(2):138-45. doi: 10.14797/mdcj-16-2-138.</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Lim</surname>
							<given-names>LJ</given-names>
						</name>
						<name>
							<surname>Tison</surname>
							<given-names>GH</given-names>
						</name>
						<name>
							<surname>Delling</surname>
							<given-names>FN</given-names>
						</name>
					</person-group>
					<article-title>Artificial Intelligence in Cardiovascular Imaging</article-title>
					<source>Methodist Debakey Cardiovasc J</source>
					<year>2020</year>
					<volume>16</volume>
					<issue>2</issue>
					<fpage>138</fpage>
					<lpage>145</lpage>
					<pub-id pub-id-type="doi">10.14797/mdcj-16-2-138</pub-id>
				</element-citation>
			</ref>
			<ref id="B22">
				<label>22</label>
				<mixed-citation>22. Celi LA, Cellini J, Charpignon ML, Dee EC, Dernoncourt F, Eber R, et al. Sources of Bias in Artificial Intelligence that Perpetuate Healthcare Disparities-A Global Review. PLOS Digit Health. 2022;1(3):e0000022. doi: 10.1371/journal.pdig.0000022.</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Celi</surname>
							<given-names>LA</given-names>
						</name>
						<name>
							<surname>Cellini</surname>
							<given-names>J</given-names>
						</name>
						<name>
							<surname>Charpignon</surname>
							<given-names>ML</given-names>
						</name>
						<name>
							<surname>Dee</surname>
							<given-names>EC</given-names>
						</name>
						<name>
							<surname>Dernoncourt</surname>
							<given-names>F</given-names>
						</name>
						<name>
							<surname>Eber</surname>
							<given-names>R</given-names>
						</name>
						<etal>et al</etal>
					</person-group>
					<article-title>Sources of Bias in Artificial Intelligence that Perpetuate Healthcare Disparities-A Global Review</article-title>
					<source>PLOS Digit Health</source>
					<year>2022</year>
					<volume>1</volume>
					<issue>3</issue>
					<elocation-id>e0000022</elocation-id>
					<pub-id pub-id-type="doi">10.1371/journal.pdig.0000022</pub-id>
				</element-citation>
			</ref>
			<ref id="B23">
				<label>23</label>
				<mixed-citation>23. Topol EJ. High-Performance Medicine: The Convergence of Human and Artificial Intelligence. Nat Med. 2019;25(1):44-56. doi: 10.1038/s41591-018-0300-7.</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Topol</surname>
							<given-names>EJ</given-names>
						</name>
					</person-group>
					<article-title>High-Performance Medicine: The Convergence of Human and Artificial Intelligence</article-title>
					<source>Nat Med</source>
					<year>2019</year>
					<volume>25</volume>
					<issue>1</issue>
					<fpage>44</fpage>
					<lpage>56</lpage>
					<pub-id pub-id-type="doi">10.1038/s41591-018-0300-7</pub-id>
				</element-citation>
			</ref>
			<ref id="B24">
				<label>24</label>
				<mixed-citation>24. Górriz JM, Ramírez J, Ortíz A, Martinez-Murcia FJ, Segovia F, Suckling J, et al. Artificial Intelligence within the Interplay between Natural and Artificial Computation: Advances in Data Science, Trends and Applications. Neurocomputing. 2020;410:237-70. doi: 10.1016/j.neucom.2020.05.078.</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Górriz</surname>
							<given-names>JM</given-names>
						</name>
						<name>
							<surname>Ramírez</surname>
							<given-names>J</given-names>
						</name>
						<name>
							<surname>Ortíz</surname>
							<given-names>A</given-names>
						</name>
						<name>
							<surname>Martinez-Murcia</surname>
							<given-names>FJ</given-names>
						</name>
						<name>
							<surname>Segovia</surname>
							<given-names>F</given-names>
						</name>
						<name>
							<surname>Suckling</surname>
							<given-names>J</given-names>
						</name>
						<etal>et al</etal>
					</person-group>
					<article-title>Artificial Intelligence within the Interplay between Natural and Artificial Computation: Advances in Data Science, Trends and Applications</article-title>
					<source>Neurocomputing</source>
					<year>2020</year>
					<volume>410</volume>
					<fpage>237</fpage>
					<lpage>270</lpage>
					<pub-id pub-id-type="doi">10.1016/j.neucom.2020.05.078</pub-id>
				</element-citation>
			</ref>
			<ref id="B25">
				<label>25</label>
				<mixed-citation>25. Alizadehsani R, Abdar M, Roshanzamir M, Khosravi A, Kebria PM, Khozeimeh F, et al. Machine Learning-Based Coronary Artery Disease Diagnosis: A Comprehensive Review. Comput Biol Med. 2019;111:103346. doi: 10.1016/j.compbiomed.2019.103346.</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Alizadehsani</surname>
							<given-names>R</given-names>
						</name>
						<name>
							<surname>Abdar</surname>
							<given-names>M</given-names>
						</name>
						<name>
							<surname>Roshanzamir</surname>
							<given-names>M</given-names>
						</name>
						<name>
							<surname>Khosravi</surname>
							<given-names>A</given-names>
						</name>
						<name>
							<surname>Kebria</surname>
							<given-names>PM</given-names>
						</name>
						<name>
							<surname>Khozeimeh</surname>
							<given-names>F</given-names>
						</name>
						<etal>et al</etal>
					</person-group>
					<article-title>Machine Learning-Based Coronary Artery Disease Diagnosis: A Comprehensive Review</article-title>
					<source>Comput Biol Med</source>
					<year>2019</year>
					<volume>111</volume>
					<size units="pages">103346</size>
					<pub-id pub-id-type="doi">10.1016/j.compbiomed.2019.103346</pub-id>
				</element-citation>
			</ref>
			<ref id="B26">
				<label>26</label>
				<mixed-citation>26. Morales MA, Manning WJ, Nezafat R. Present and Future Innovations in AI and Cardiac MRI. Radiology. 2024;310(1):e231269. doi: 10.1148/radiol.231269.</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Morales</surname>
							<given-names>MA</given-names>
						</name>
						<name>
							<surname>Manning</surname>
							<given-names>WJ</given-names>
						</name>
						<name>
							<surname>Nezafat</surname>
							<given-names>R</given-names>
						</name>
					</person-group>
					<article-title>Present and Future Innovations in AI and Cardiac MRI</article-title>
					<source>Radiology</source>
					<year>2024</year>
					<volume>310</volume>
					<issue>1</issue>
					<elocation-id>e231269</elocation-id>
					<pub-id pub-id-type="doi">10.1148/radiol.231269</pub-id>
				</element-citation>
			</ref>
			<ref id="B27">
				<label>27</label>
				<mixed-citation>27. Echefu G, Shah R, Sanchez Z, Rickards J, Brown SA. Artificial Intelligence: Applications in Cardio-Oncology and Potential Impact on Racial Disparities. Am Heart J Plus. 2024;48:100479. doi: 10.1016/j.ahjo.2024.100479.</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Echefu</surname>
							<given-names>G</given-names>
						</name>
						<name>
							<surname>Shah</surname>
							<given-names>R</given-names>
						</name>
						<name>
							<surname>Sanchez</surname>
							<given-names>Z</given-names>
						</name>
						<name>
							<surname>Rickards</surname>
							<given-names>J</given-names>
						</name>
						<name>
							<surname>Brown</surname>
							<given-names>SA</given-names>
						</name>
					</person-group>
					<article-title>Artificial Intelligence: Applications in Cardio-Oncology and Potential Impact on Racial Disparities</article-title>
					<source>Am Heart J Plus</source>
					<year>2024</year>
					<volume>48</volume>
					<size units="pages">100479</size>
					<pub-id pub-id-type="doi">10.1016/j.ahjo.2024.100479</pub-id>
				</element-citation>
			</ref>
			<ref id="B28">
				<label>28</label>
				<mixed-citation>28. Chen RJ, Lu MY, Chen TY, Williamson DFK, Mahmood F. Synthetic Data in Machine Learning for Medicine and Healthcare. Nat Biomed Eng. 2021;5(6):493-7. doi: 10.1038/s41551-021-00751-8.</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Chen</surname>
							<given-names>RJ</given-names>
						</name>
						<name>
							<surname>Lu</surname>
							<given-names>MY</given-names>
						</name>
						<name>
							<surname>Chen</surname>
							<given-names>TY</given-names>
						</name>
						<name>
							<surname>Williamson</surname>
							<given-names>DFK</given-names>
						</name>
						<name>
							<surname>Mahmood</surname>
							<given-names>F</given-names>
						</name>
					</person-group>
					<article-title>Synthetic Data in Machine Learning for Medicine and Healthcare</article-title>
					<source>Nat Biomed Eng</source>
					<year>2021</year>
					<volume>5</volume>
					<issue>6</issue>
					<fpage>493</fpage>
					<lpage>497</lpage>
					<pub-id pub-id-type="doi">10.1038/s41551-021-00751-8</pub-id>
				</element-citation>
			</ref>
			<ref id="B29">
				<label>29</label>
				<mixed-citation>29. Sengupta PP, Shrestha S, Berthon B, Messas E, Donal E, Tison GH, et al. Proposed Requirements for Cardiovascular Imaging-Related Machine Learning Evaluation (PRIME): A Checklist: Reviewed by the American College of Cardiology Healthcare Innovation Council. JACC Cardiovasc Imaging. 2020;13(9):2017-35. doi: 10.1016/j.jcmg.2020.07.015.</mixed-citation>
				<element-citation publication-type="journal">
					<person-group person-group-type="author">
						<name>
							<surname>Sengupta</surname>
							<given-names>PP</given-names>
						</name>
						<name>
							<surname>Shrestha</surname>
							<given-names>S</given-names>
						</name>
						<name>
							<surname>Berthon</surname>
							<given-names>B</given-names>
						</name>
						<name>
							<surname>Messas</surname>
							<given-names>E</given-names>
						</name>
						<name>
							<surname>Donal</surname>
							<given-names>E</given-names>
						</name>
						<name>
							<surname>Tison</surname>
							<given-names>GH</given-names>
						</name>
						<etal>et al</etal>
					</person-group>
					<article-title>Proposed Requirements for Cardiovascular Imaging-Related Machine Learning Evaluation (PRIME): A Checklist: Reviewed by the American College of Cardiology Healthcare Innovation Council</article-title>
					<source>JACC Cardiovasc Imaging</source>
					<year>2020</year>
					<volume>13</volume>
					<issue>9</issue>
					<fpage>2017</fpage>
					<lpage>2035</lpage>
					<pub-id pub-id-type="doi">10.1016/j.jcmg.2020.07.015</pub-id>
				</element-citation>
			</ref>
		</ref-list>
		<fn-group>
			<fn fn-type="other">
				<label>Study Association:</label>
				<p> This study is not associated with any thesis or dissertation work.</p>
			</fn>
			<fn fn-type="other">
				<label>Ethics Approval and Consent to Participate:</label>
				<p> This article does not contain any studies with human participants or animals performed by any of the authors.</p>
			</fn>
			<fn fn-type="other">
				<label>Use of Artificial Intelligence:</label>
				<p> The authors did not use any artificial intelligence tools in the development of this work.</p>
			</fn>
			<fn fn-type="data-availability" specific-use="data-available-upon-request">
				<label>Availability of Research Data:</label>
				<p> All datasets supporting the results of this study are available upon request from the corresponding author.</p>
			</fn>
			<fn fn-type="financial-disclosure">
				<label>Sources of Funding:</label>
				<p> There were no external funding sources for this study.</p>
			</fn>
		</fn-group>
	</back>
	<sub-article article-type="translation" id="TRpt" xml:lang="pt">
		<front-stub>
			<article-id pub-id-type="doi">10.36660/abchf.20260016</article-id>
			<article-categories>
				<subj-group subj-group-type="heading">
					<subject>Artigo Original</subject>
				</subj-group>
			</article-categories>
			<title-group>
				<article-title>Mapeamento Global da Pesquisa Sobre Inteligência Artificial Aplicada à Miocardite</article-title>
			</title-group>
			<contrib-group>
				<contrib contrib-type="author">
					<contrib-id contrib-id-type="orcid">0000-0002-2386-048X</contrib-id>
					<name>
						<surname>Costa</surname>
						<given-names>Guilherme Cesar Fernandes de Oliveira da</given-names>
					</name>
					<role> Concepção e desenho da pesquisa e Análise estatística</role>
					<role>Obtenção de dados</role>
					<role>Análise e interpretação dos dados</role>
					<role>Redação do manuscrito e Revisão crítica do manuscrito quanto ao conteúdo intelectual importante</role>
					<xref ref-type="aff" rid="aff1002"><sup>1</sup></xref>
				</contrib>
				<contrib contrib-type="author">
					<contrib-id contrib-id-type="orcid">0009-0004-7596-1014</contrib-id>
					<name>
						<surname>Rodriguez</surname>
						<given-names>Jose Gregorio Valero</given-names>
					</name>
					<role> Concepção e desenho da pesquisa e Análise estatística</role>
					<role>Análise e interpretação dos dados</role>
					<role>Redação do manuscrito e Revisão crítica do manuscrito quanto ao conteúdo intelectual importante</role>
					<xref ref-type="aff" rid="aff1002"><sup>1</sup></xref>
				</contrib>
				<contrib contrib-type="author">
					<contrib-id contrib-id-type="orcid">0009-0005-6008-4335</contrib-id>
					<name>
						<surname>Costa</surname>
						<given-names>Giulia Fernandes de Oliveira da</given-names>
					</name>
					<role>Concepção e desenho da pesquisa e Análise estatística</role>
					<role>Obtenção de dados</role>
					<role>Análise e interpretação dos dados</role>
					<role>Redação do manuscrito e Revisão crítica do manuscrito quanto ao conteúdo intelectual importante</role>
					<xref ref-type="aff" rid="aff2002"><sup>2</sup></xref>
				</contrib>
				<contrib contrib-type="author">
					<contrib-id contrib-id-type="orcid">0000-0002-8870-5130</contrib-id>
					<name>
						<surname>Pimentel</surname>
						<given-names>Adriana Munford Lima</given-names>
					</name>
					<role>Análise e interpretação dos dados</role>
					<role>Redação do manuscrito e Revisão crítica do manuscrito quanto ao conteúdo intelectual importante</role>
					<xref ref-type="aff" rid="aff3002"><sup>3</sup></xref>
				</contrib>
				<contrib contrib-type="author">
					<contrib-id contrib-id-type="orcid">0000-0002-7452-3870</contrib-id>
					<name>
						<surname>Mesquita</surname>
						<given-names>Evandro Tinoco</given-names>
					</name>
					<role>Análise e interpretação dos dados</role>
					<role>Redação do manuscrito e Revisão crítica do manuscrito quanto ao conteúdo intelectual importante</role>
					<xref ref-type="aff" rid="aff3002"><sup>3</sup></xref>
				</contrib>
			</contrib-group>
			<aff id="aff1002">
				<label>1</label>
				<country country="BR">Brasil</country>
				<institution content-type="original">Universidade Federal Fluminense, Niterói, RJ – Brasil</institution>
			</aff>
			<aff id="aff2002">
				<label>2</label>
				<country country="BR">Brasil</country>
				<institution content-type="original">Faculdade de Ciências Médicas de Maricá – Faculdade de Medicina, Maricá, RJ – Brasil</institution>
			</aff>
			<aff id="aff3002">
				<label>3 </label>
				<country country="BR">Brasil</country>
				<institution content-type="original">Universidade Federal Fluminense – Hospital Universitário Antônio Pedro,Niterói, RJ – Brasil</institution>
			</aff>
			<author-notes>
				<corresp id="c01002">
					<label>Correspondência:</label> Evandro Tinoco Mesquita • Universidade Federal Fluminense – R. Des. Athayde Parreiras, 100. CEP 24070-090, Fátima, Niterói, RJ – Brasil E-mail: etmesquita@gmail.com </corresp>
				<fn fn-type="edited-by">
					<label>Editor responsável pela revisão:</label>
					<p>Victor Issa</p>
				</fn>
				<fn fn-type="coi-statement">
					<label>Potencial Conflito de Interesse:</label>
					<p> Declaro não haver conflito de interesses pertinentes.</p>
				</fn>
			</author-notes>
			<abstract>
				<title>Resumo</title>
				<sec>
					<title>Fundamento</title>
					<p> A miocardite continua sendo um desafio diagnóstico, e o interesse no uso da inteligência artificial para apoiar a avaliação diagnóstica não invasiva cresceu rapidamente. Embora a literatura na interseção entre inteligência artificial e miocardite tenha se expandido substancialmente, ela permanece fragmentada e carece de uma síntese abrangente.</p>
				</sec>
				<sec>
					<title>Objetivos</title>
					<p> Mapear a produção científica global sobre a aplicação da inteligência artificial à miocardite, identificando os principais contribuintes, padrões de colaboração e tendências temáticas emergentes.</p>
				</sec>
				<sec>
					<title>Métodos</title>
					<p> Realizamos uma análise bibliométrica de artigos sobre inteligência artificial aplicada à miocardite indexados na Web of Science Core Collection. O pacote <italic>bibliometrix</italic> para R e o VOSviewer foram utilizados para realizar análises de redes bibliométricas e análises temáticas.</p>
				</sec>
				<sec>
					<title>Resultados</title>
					<p> Foram incluídos 47 artigos. A produção científica aumentou acentuadamente após 2021, coincidindo com a pandemia de covid-19, e atingiu seu pico em 2025. A China liderou em volume de publicações, enquanto o Reino Unido e a Alemanha alcançaram o maior impacto de citações. Nenhuma publicação teve origem na América Latina, no Caribe ou na África. As redes de colaboração internacional permanecem fragmentadas. As análises de palavras-chave e temáticas identificaram a integração do aprendizado profundo com a ressonância magnética cardiovascular como o principal tema motor, refletindo uma mudança em relação à tradicional biópsia endomiocárdica. Os autores mais citados eram predominantemente cientistas da computação e da ciência de dados.</p>
				</sec>
				<sec>
					<title>Conclusões</title>
					<p> A pesquisa na interseção entre inteligência artificial e miocardite constitui um campo emergente em processo de consolidação metodológica, com crescimento acelerado pela pandemia de covid-19. A ressonância magnética cardiovascular tornou-se a principal fonte de dados para aplicações baseadas em inteligência artificial. O principal desafio parece ser translacional, ressaltando a necessidade de uma colaboração mais estreita entre desenvolvedores de algoritmos e clínicos. A integração bem-sucedida desses modelos à prática clínica rotineira provavelmente dependerá da colaboração multidisciplinar e da validação em populações multicêntricas e diversas.</p>
				</sec>
			</abstract>
			<kwd-group xml:lang="pt">
				<title>Palavras-chave</title>
				<kwd>Miocardite</kwd>
				<kwd>Inteligência Artificial</kwd>
				<kwd>Aprendizado de Máquina</kwd>
				<kwd>Aprendizagem Profunda</kwd>
			</kwd-group>
			<funding-group>
				<funding-statement><bold>Fontes de Financiamento:</bold> O presente estudo não teve fontes de financiamento externas</funding-statement>
			</funding-group>
		</front-stub>
		<body>
			<p>
					<fig id="f01002">
						<label>Figura Central:</label>
						<caption>
							<title>Mapeamento Global da Pesquisa Sobre Inteligência Artificial Aplicada à Miocardite</title>
						</caption>
						<graphic xlink:href="2764-3107-abchf-6-e20260016-gf01-pt.tif"/>
						<attrib><italic>Síntese bibliométrica descritiva de 47 publicações (2019-2026) indexadas na Web of Science Core Collection, ilustrando a produção científica anual, a estrutura temática do campo (centrada na RMC), a distribuição geográfica e o perfil interdisciplinar de autoria. IA: inteligência artificial; GAN: rede adversarial generativa; PM: publicação multinacional; RMC: ressonância magnética cardiovascular.</italic></attrib>
					</fig>
				</p>
			<sec sec-type="intro">
				<title>Introdução</title>
				<p>A miocardite é caracterizada por inflamação difusa ou focal do miocárdio, resultante, mais comumente, de agentes infecciosos, reações imunomediadas ou processos inflamatórios crônicos.<sup><xref ref-type="bibr" rid="B1">1</xref></sup> Clinicamente, continua sendo um desafio diagnóstico devido à sua apresentação heterogênea e frequentemente inespecífica, variando de um curso assintomático à insuficiência cardíaca, arritmias e morte súbita.<sup><xref ref-type="bibr" rid="B1">1</xref></sup> Tradicionalmente, a biópsia endomiocárdica tem sido considerada o padrão-ouro diagnóstico; entretanto, os avanços na ressonância magnética cardiovascular e o refinamento dos critérios de Lake Louise transformaram a abordagem diagnóstica não invasiva.<sup><xref ref-type="bibr" rid="B1">1</xref></sup> Cenários clínicos emergentes, particularmente a expansão da cardio-oncologia impulsionada pelo aumento da incidência de miocardite associada aos inibidores de <italic>checkpoint</italic> imunológico, intensificaram ainda mais a necessidade de ferramentas diagnósticas mais precisas e oportunas, capazes de apoiar a tomada de decisão clínica, ao mesmo tempo em que reduzem a dependência de procedimentos invasivos.<sup><xref ref-type="bibr" rid="B2">2</xref></sup></p>
				<p>A inteligência artificial, incluindo os subcampos do aprendizado de máquina e do aprendizado profundo, emerge como uma ferramenta promissora para o apoio diagnóstico.<sup><xref ref-type="bibr" rid="B3">3</xref></sup> A capacidade desses algoritmos de processar grandes volumes de dados, identificar padrões fenotípicos complexos e predizer desfechos clínicos sugere que a inteligência artificial pode impulsionar a medicina de precisão e contribuir para a evolução da cardiologia moderna.<sup><xref ref-type="bibr" rid="B3">3</xref></sup> Apesar do crescente interesse tecnológico, a pesquisa na interseção entre inteligência artificial e miocardite expandiu-se de forma fragmentada, com esforços limitados para caracterizar sistematicamente o campo, identificar as principais áreas de pesquisa, mapear redes de colaboração ou destacar tendências metodológicas emergentes.</p>
				<p>A análise bibliométrica fornece uma estrutura metodológica robusta para mapear o panorama científico de um campo de pesquisa.<sup><xref ref-type="bibr" rid="B4">4</xref></sup> Diferentemente das revisões sistemáticas convencionais, as abordagens bibliométricas permitem a visualização da estrutura intelectual de um campo, a identificação de redes de coautoria, a avaliação da distribuição geográfica da produção científica e a análise da evolução temporal dos tópicos de pesquisa.<sup><xref ref-type="bibr" rid="B4">4</xref>,<xref ref-type="bibr" rid="B5">5</xref></sup></p>
				<p>Este estudo teve como objetivo realizar uma análise bibliométrica da literatura científica global sobre a aplicação da inteligência artificial à miocardite. Especificamente, buscamos identificar os principais contribuintes (i.e., autores, países, instituições e periódicos), bem como os principais agrupamentos temáticos do campo, fornecendo, assim, uma referência abrangente para orientar pesquisas futuras e apoiar a integração das tecnologias de inteligência artificial à medicina cardiovascular personalizada (<xref ref-type="fig" rid="f01002">Figura Central</xref>).</p>
			</sec>
			<sec sec-type="methods">
				<title>Métodos</title>
				<sec>
					<title>Fonte de dados</title>
					<p>A Web of Science Core Collection foi selecionada como fonte de dados para esta análise bibliométrica devido à sua ampla cobertura de periódicos científicos indexados.<sup><xref ref-type="bibr" rid="B6">6</xref></sup> A seguinte <italic>string</italic> de busca avançada foi aplicada: (“<italic>myocarditis</italic>” OR “<italic>inflammatory cardiomyopathy</italic>”) AND (“<italic>artificial intelligence</italic>” OR “<italic>machine learning</italic>” OR “<italic>deep learning</italic>”). A busca foi realizada em 27 de fevereiro de 2026. Foram incluídos apenas artigos de pesquisa originais, sem restrições quanto ao ano de publicação.</p>
					<p>Todos os registros elegíveis foram recuperados em uma única sessão de busca, resultando em 82 artigos. Dois revisores independentes realizaram a triagem manual dos títulos e resumos para excluir registros duplicados e estudos cujo foco principal não fosse a miocardite, seguindo recomendações padronizadas para pesquisas bibliométricas na literatura médica.<sup><xref ref-type="bibr" rid="B7">7</xref></sup> Esse processo de triagem resultou em um conjunto final de dados composto por 47 artigos para análise. Os registros bibliográficos foram exportados em formato de texto simples e salvos como arquivos download.txt para processamento subsequente.</p>
				</sec>
				<sec>
					<title>Análise bibliométrica</title>
					<p>O processamento dos dados, as análises bibliométricas e as visualizações foram realizados utilizando o R (versão 4.5.2) e o VOSviewer (versão 1.6.20). O ambiente R, por meio do pacote <italic>bibliometrix</italic>, possibilitou análises bibliométricas abrangentes, incluindo extração de dados, construção de redes e avaliação da produtividade científica e do impacto de países, instituições, periódicos, autores e referências cocitadas.<sup><xref ref-type="bibr" rid="B5">5</xref></sup> Os resultados foram apresentados por meio de mapas de redes bibliométricas e visualizações gráficas.</p>
					<p>Também foram realizadas análises temporais para examinar a evolução das palavras-chave de alta frequência, fornecendo informações sobre o desenvolvimento dos tópicos de pesquisa e das tendências emergentes no campo. Além disso, o VOSviewer foi utilizado para construir e visualizar redes de cocitação e de coocorrência de palavras-chave.<sup><xref ref-type="bibr" rid="B8">8</xref></sup></p>
				</sec>
			</sec>
			<sec sec-type="results">
				<title>Resultados</title>
				<sec>
					<title>Análise das publicações</title>
					<p>Um total de 47 publicações, publicadas entre 2019 e 2026, foi incluído neste estudo. A distribuição temporal das publicações demonstra uma produção científica modesta até 2020, seguida por um crescimento sustentado a partir de 2021. O volume anual de publicações aumentou de dois artigos em 2019 para 13 em 2025, representando o maior nível de produção científica durante o período do estudo. O aparente declínio observado em 2026 reflete a cobertura parcial do ano de publicação, uma vez que a busca na literatura foi realizada em 27 de fevereiro de 2026.</p>
					<p>Os padrões de citação diferiram das tendências de publicação. O impacto científico, mensurado pelo número médio de citações por artigo por ano, atingiu seu pico em 2021, com 5,7 citações/ano. Posteriormente, a taxa média anual de citações diminuiu progressivamente, alcançando valores próximos de zero para os artigos publicados em 2025 e 2026, refletindo o tempo limitado disponível para que estudos recém-publicados acumulassem citações (<xref ref-type="fig" rid="f02002">Figura 1A</xref>).</p>
					<p>
						<fig id="f02002">
							<label>Figura 1</label>
							<caption>
								<title>– Produção científica e distribuição geográfica. A) Produção científica anual e média de citações totais por artigo. B) Produção científica por país, classificada em PNs e PMs. PN: publicação nacional; PM: publicação multinacional.</title>
							</caption>
							<graphic xlink:href="2764-3107-abchf-6-e20260016-gf02-pt.tif"/>
						</fig>
					</p>
				</sec>
				<sec>
					<title>Análise dos países</title>
					<p>A análise das afiliações dos autores correspondentes mostrou que a China foi o principal contribuinte (n = 15 [31,9%]). A Alemanha ocupou a segunda posição (n = 7 [14,9%]), seguida pelos Estados Unidos (n = 6 [12,8%]) e pelo Reino Unido (n = 4 [8,5%]).</p>
					<p>Os padrões de colaboração internacional variaram consideravelmente entre os países mais produtivos. A Alemanha apresentou a maior proporção de publicações multinacionais (PM), com 57,1% de seus estudos (quatro de sete) envolvendo colaboração internacional. Em contraste, a produção científica da China foi predominantemente nacional, compreendendo 11 publicações nacionais e uma taxa de PM de 26,7%. Entre os principais países em número de publicações, os Estados Unidos apresentaram o menor nível de colaboração internacional, com cinco de suas seis publicações envolvendo apenas autores nacionais (taxa de PM: 16,7%) (<xref ref-type="fig" rid="f02002">Figura 1B</xref>).</p>
					<p>Em termos de impacto científico, o Reino Unido ocupou a primeira posição, com 152 citações totais e uma média de 38 citações por artigo. A China ficou em segundo lugar em número total de citações (76), embora sua taxa média de citações tenha sido substancialmente menor (5,1 citações/artigo). A Alemanha acumulou 63 citações no total, correspondendo a uma média de nove citações por publicação (<xref ref-type="table" rid="t1002">Tabela 1</xref>).</p>
					<p>
						<table-wrap id="t1002">
							<label>Tabela 1</label>
							<caption>
								<title>– Líderes globais em produtividade científica na pesquisa sobre inteligência artificial aplicada à miocardite</title>
							</caption>
							<table frame="hsides" rules="groups">
								<colgroup>
									<col/>
									<col/>
									<col/>
									<col/>
								</colgroup>
								<thead>
									<tr>
										<th align="left">Classificação</th>
										<th>Principais países (citações totais)</th>
										<th>Principais instituições (total de artigos)</th>
										<th>Periódicos mais citados (citações totais)</th>
									</tr>
								</thead>
								<tbody>
									<tr>
										<td>1</td>
										<td align="center">Reino Unido (152)</td>
										<td align="center">University of Bonn (11)</td>
										<td align="center">Circulation (58)</td>
									</tr>
									<tr>
										<td>2</td>
										<td align="center">China (76)</td>
										<td align="center">Taipei Medical University (10)</td>
										<td align="center">Journal of the American College of Cardiology (53)</td>
									</tr>
									<tr>
										<td>3</td>
										<td align="center">Alemanha (63)</td>
										<td align="center">University of London (8)</td>
										<td align="center">Journal of Cardiovascular Magnetic Resonance (42)</td>
									</tr>
									<tr>
										<td>4</td>
										<td align="center">Suíça (36)</td>
										<td align="center">Mayo Clinic (7)</td>
										<td align="center">Radiology (42)</td>
									</tr>
									<tr>
										<td>5</td>
										<td align="center">Estados Unidos (36)</td>
										<td align="center">Technical University of Munich (6)</td>
										<td align="center">European Heart Journal (34)</td>
									</tr>
									<tr>
										<td>6</td>
										<td align="center">Austrália (32)</td>
										<td align="center">Imperial College London (5)</td>
										<td align="center">JACC: Cardiovascular Imaging (33)</td>
									</tr>
									<tr>
										<td>7</td>
										<td align="center">Itália (16)</td>
										<td align="center">Deakin University (4)</td>
										<td align="center">Lecture Notes in Computer Science (33)</td>
									</tr>
									<tr>
										<td>8</td>
										<td align="center">Áustria (8)</td>
										<td align="center">Johannes Gutenberg University Mainz (4)</td>
										<td align="center">European Radiology (25)</td>
									</tr>
									<tr>
										<td>9</td>
										<td align="center">Irã (8)</td>
										<td align="center">Medical University of Innsbruck (4)</td>
										<td align="center">Scientific Reports (22)</td>
									</tr>
									<tr>
										<td>10</td>
										<td align="center">Coreia (4)</td>
										<td align="center">Okayama University (4)</td>
										<td align="center">arXiv (21)</td>
									</tr>
								</tbody>
							</table>
							<table-wrap-foot>
								<fn id="TFN1002">
									<p><italic>Nota: Os dados representam as 10 principais entidades em cada categoria com base nas publicações indexadas na Web of Science Core Collection (2019-2026). Os valores entre parênteses indicam o número total de citações para países e periódicos e o número total de artigos originais para as instituições.</italic></p>
								</fn>
							</table-wrap-foot>
						</table-wrap>
					</p>
					<p>Apesar de produzir menos publicações, a Suíça e a Austrália alcançaram elevado impacto de citações, com médias de 36 e 32 citações por artigo, respectivamente. Os Estados Unidos acumularam 36 citações totais, correspondendo a uma média de seis citações por publicação. De modo geral, a produção científica concentrou-se na China, na Europa e nos Estados Unidos.</p>
				</sec>
				<sec>
					<title>Análise das afiliações</title>
					<p>As instituições com maior produção científica são apresentadas na <xref ref-type="table" rid="t1002">Tabela 1</xref>. A University of Bonn ocupou a primeira posição, com 11 publicações, representando aproximadamente 25% da amostra total, seguida pela Taipei Medical University (n = 10), University of London (n = 8) e Mayo Clinic (n = 7).</p>
					<p>A colaboração interinstitucional foi caracterizada por redes de pesquisa compartimentalizadas, compostas por agrupamentos com interação limitada. O agrupamento laranja representa uma rede de colaboração euro-australiana envolvendo a Deakin University e a University of Granada, destacando a contribuição de grupos de pesquisa especializados em ciência da computação e processamento de sinais (<xref ref-type="fig" rid="f03002">Figura 2A</xref>).</p>
					<p>
						<fig id="f03002">
							<label>Figura 2</label>
							<caption>
								<title>– Redes de colaboração institucional e entre autores. A) Rede de colaboração institucional. B) Rede de colaboração entre autores.</title>
							</caption>
							<graphic xlink:href="2764-3107-abchf-6-e20260016-gf03-pt.tif"/>
						</fig>
					</p>
					<p>A rede alemã, representada pelo agrupamento vermelho, caracteriza-se por uma intensa colaboração regional entre centros médicos, liderada pela Johannes Gutenberg University Mainz, University of Bonn e Hannover Medical School. O agrupamento verde representa o consórcio de pesquisa britânico, compreendendo fortes vínculos colaborativos entre a University of London, Imperial College London e King’s College London.</p>
				</sec>
				<sec>
					<title>Análise dos autores</title>
					<p>A produtividade dos autores foi liderada por Alizadehsani R (n = 4), seguido por Górriz JM (n = 3). Em termos de impacto local, definido como o número de vezes que um documento dentro do conjunto de dados analisado foi citado por outros documentos do mesmo conjunto de dados, oito autores empataram com a maior pontuação, cada um com quatro citações locais: Alizadehsani R, Alkadhi H, Di Noto T, Gantert E, Manka R, Mannil M, Soda P e von Spiczak J.</p>
					<p>A aplicação da Lei de Lotka demonstrou um desvio acentuado em relação à distribuição teórica clássica.<sup><xref ref-type="bibr" rid="B9">9</xref></sup> No geral, 98,4% dos autores contribuíram com apenas uma única publicação, enquanto apenas uma pequena proporção foi autora de dois ou mais artigos.</p>
					<p>A rede de coautoria na pesquisa sobre inteligência artificial aplicada à miocardite foi fragmentada em oito agrupamentos distintos, dois dos quais estavam centrados nos autores com maior impacto local. O maior agrupamento (vermelho) foi organizado em torno de Alizadehsani R e Górriz JM, enquanto o agrupamento azul compreendeu Kaya K, Luetkens JA, Pennig L e Zimmer S. A visualização da rede evidencia a presença de grupos colaborativos isolados, sem conexões externas, sendo a espessura das ligações representativa da frequência de coautoria direta (<xref ref-type="fig" rid="f03002">Figura 2B</xref>).</p>
					<p>Indicadores adicionais de impacto científico local, incluindo o índice H e o índice M, caracterizaram ainda mais a influência dos autores no campo. Alizadehsani R alcançou o maior índice H (H = 4), com quatro publicações recebendo pelo menos quatro citações cada. Górriz JM ocupou a segunda posição (H = 3). Após ajuste pela idade acadêmica, Górriz JM obteve o maior índice M (M = 1,0), seguido por Alizadehsani R (M = 0,8). Três autores adicionais apresentaram desempenho comparável, cada um com índice H igual a 2 e índice M igual a 0,67: Kaya K, Luetkens JA e Pennig L (<xref ref-type="fig" rid="f04002">Figura 3</xref>).</p>
					<p>
						<fig id="f04002">
							<label>Figura 3</label>
							<caption>
								<title>– Métricas de impacto dos autores.</title>
							</caption>
							<graphic xlink:href="2764-3107-abchf-6-e20260016-gf04-pt.tif"/>
						</fig>
					</p>
				</sec>
				<sec>
					<title>Análise dos periódicos</title>
					<p>A distribuição da produção científica por periódico identificou o Journal of Clinical Medicine como o periódico mais produtivo, com quatro artigos publicados. Entretanto, a análise do impacto local de citações mostrou que o periódico Circulation ocupou a primeira posição em número total de citações (n = 58), seguido pelo Journal of the American College of Cardiology (n = 53), Journal of Cardiovascular Magnetic Resonance (n = 42) e Radiology (n = 42) (<xref ref-type="table" rid="t1002">Tabela 1</xref>).</p>
					<p>A aplicação da Lei de Bradford caracterizou a dispersão das publicações e identificou os periódicos centrais deste campo de pesquisa.<sup><xref ref-type="bibr" rid="B10">10</xref></sup> A análise revelou um pequeno núcleo (Zona 1), composto por quatro periódicos que concentraram o maior número de publicações sobre o tema, cada um contribuindo com dois artigos: i) CAAI Transactions on Intelligence Technology; ii) Diagnostics; iii) Journal of Cardiovascular Magnetic Resonance; e iv) Quantitative Imaging in Medicine and Surgery.</p>
				</sec>
				<sec>
					<title>Análise das citações globais</title>
					<p>As contagens globais de citações fornecem um indicador da influência científica de um artigo na comunidade acadêmica em sentido amplo. A publicação mais citada do conjunto de dados (n = 116) foi o estudo de primeira autoria de Joy G,<sup><xref ref-type="bibr" rid="B11">11</xref></sup> intitulado “Prospective Case-Control Study of Cardiovascular Abnormalities 6 Months Following Mild COVID-19 in Healthcare Workers,” publicado no Journal of the American College of Cardiology. O estudo investigou anormalidades cardiovasculares após infecção por SARS-CoV-2 utilizando abordagens analíticas baseadas em inteligência artificial.</p>
					<p>O segundo artigo mais citado (n = 41), intitulado “CNN-KCL: Automatic myocarditis diagnosis using convolutional neural network combined with k-means clustering”, teve Sharifrazi D como primeiro autor,<sup><xref ref-type="bibr" rid="B12">12</xref></sup> com Alizadehsani R como coautor. O estudo propôs uma estrutura baseada em aprendizado profundo para o diagnóstico automatizado da miocardite.</p>
				</sec>
				<sec>
					<title>Análise das referências cocitadas</title>
					<p>A estrutura intelectual da pesquisa sobre inteligência artificial aplicada à miocardite foi mapeada por meio da análise de redes de cocitação em três níveis: i) periódicos; ii) autores; e iii) referências individuais.</p>
					<p>A rede de cocitação de periódicos apresentou uma estrutura tripartite. O agrupamento vermelho representa a cardiologia clínica e compreende periódicos de alto impacto, incluindo Circulation, Journal of the American College of Cardiology e European Heart Journal. O agrupamento azul corresponde ao domínio da imagem cardiovascular, abrangendo periódicos especializados como Radiology, Journal of Cardiovascular Magnetic Resonance e JACC: Cardiovascular Imaging. O agrupamento verde reflete a dimensão tecnológica e multidisciplinar do campo, integrando periódicos de ciência da computação e de pesquisa científica geral, com o periódico Lecture Notes in Computer Science ocupando posição de destaque (<xref ref-type="fig" rid="f04002">Figura 4A</xref>).</p>
					<p>
						<fig id="f05002">
							<label>Figura 4</label>
							<caption>
								<title>– Redes de cocitação de periódicos e autores. A) Rede de cocitação de periódicos. B) Rede de cocitação de autores.</title>
							</caption>
							<graphic xlink:href="2764-3107-abchf-6-e20260016-gf05-pt.tif"/>
						</fig>
					</p>
					<p>A rede de cocitação de autores revelou duas principais frentes intelectuais na interseção entre inteligência artificial e miocardite. O agrupamento azul representa o domínio clínico e diagnóstico, centrado em Ferreira VM, cujo trabalho sobre ressonância magnética cardiovascular tornou-se um marco no diagnóstico da inflamação miocárdica (<xref ref-type="fig" rid="f04002">Figura 4B</xref>).<sup><xref ref-type="bibr" rid="B13">13</xref></sup></p>
					<p>O agrupamento vermelho representa o domínio algorítmico e metodológico, centrado em autores cujos trabalhos se concentram no desenvolvimento de modelos computacionais, incluindo Moravvej SV,<sup><xref ref-type="bibr" rid="B14">14</xref></sup> Vakilian S<sup><xref ref-type="bibr" rid="B15">15</xref></sup> e Hong L.<sup><xref ref-type="bibr" rid="B16">16</xref></sup> Esse agrupamento também inclui importantes cientistas da computação, como Cortes C, codesenvolvedor das máquinas de vetores de suporte (SVMs).<sup><xref ref-type="bibr" rid="B17">17</xref></sup> A rede demonstra uma forte conexão estrutural entre esses dois domínios, refletindo a frequente cocitação de autores das ciências médicas e computacionais.</p>
					<p>No nível das referências, a rede de cocitação posiciona o artigo de Ferreira VM (2018) em seu centro, conectando múltiplos agrupamentos temáticos. Essa publicação seminal introduziu a atualização dos critérios de Lake Louise para o diagnóstico da miocardite por meio da ressonância magnética cardiovascular.<sup><xref ref-type="bibr" rid="B13">13</xref></sup> A rede também destaca contribuições metodológicas fundamentais, incluindo o trabalho de Cortes C,<sup><xref ref-type="bibr" rid="B17">17</xref></sup> que estabeleceu as bases teóricas das SVMs. Com base nesses estudos fundamentais, emergiu um conjunto denso e altamente interconectado de pesquisas aplicadas recentes, com contribuições notáveis de Moravvej SV (2021, 2022), Danaei S (2022), Sudre CH (2017) e Hong L (2023) (<xref ref-type="fig" rid="f06002">Figura 5</xref>).</p>
					<p>
						<fig id="f06002">
							<label>Figura 5</label>
							<caption>
								<title>– Rede de cocitação de referências.</title>
							</caption>
							<graphic xlink:href="2764-3107-abchf-6-e20260016-gf06-pt.tif"/>
						</fig>
					</p>
				</sec>
				<sec>
					<title>Análise de coocorrência</title>
					<p>A estrutura conceitual do campo foi mapeada por meio da análise de coocorrência de palavras-chave, identificando termos que coocorreram frequentemente entre os artigos incluídos no conjunto de dados. Três palavras-chave de alta frequência (<italic>myocarditis, diagnosis</italic> e <italic>artificial intelligence</italic>) serviram como nós centrais dos principais agrupamentos temáticos (<xref ref-type="fig" rid="f07002">Figura 6</xref>).</p>
					<p>
						<fig id="f07002">
							<label>Figura 6</label>
							<caption>
								<title>– Rede de coocorrência de palavras-chave.</title>
							</caption>
							<graphic xlink:href="2764-3107-abchf-6-e20260016-gf07-pt.tif"/>
						</fig>
					</p>
					<p>O agrupamento vermelho está centrado em <italic>myocarditis</italic> e relaciona cenários clínicos desafiadores, incluindo miocardite fulminante e miocardite associada a inibidores de <italic>checkpoint</italic> imunológico, a abordagens computacionais avançadas, como redes adversariais generativas, aprendizado por reforço e técnicas de aumento de dados.</p>
					<p>O agrupamento azul está organizado em torno de <italic>diagnosis</italic> e compreende termos relacionados ao desenvolvimento de algoritmos, incluindo <italic>deep learning, classification</italic> e <italic>prediction</italic>, que estão fortemente associados à ressonância magnética cardiovascular. Os agrupamentos verde e roxo conectam <italic>artificial intelligence</italic> e <italic>machine learning</italic> a diagnósticos diferenciais e biomarcadores de imagem, incluindo sarcoidose cardíaca, miocardite viral e realce tardio pelo gadolínio.</p>
					<p>Os agrupamentos laranja e marrom sugerem a influência substancial da pandemia de covid-19 sobre a evolução da literatura. O agrupamento laranja está centrado em <italic>COVID-19</italic> e <italic>SARS-CoV-2</italic>, enquanto os agrupamentos marrom e roxo refletem a influência das recomendações de sociedades profissionais, representadas por termos como <italic>position statement</italic> e <italic>working group</italic>, destacando os esforços em andamento para padronizar o diagnóstico da miocardite. Quando a rede foi filtrada para reter apenas as associações mais fortes, a relação mais proeminente e densamente conectada foi observada entre os nós <italic>diagnosis, cardiovascular magnetic resonance imaging</italic> e <italic>myocarditis</italic>.</p>
				</sec>
				<sec>
					<title>Hotspots de pesquisa e tendências emergentes</title>
					<p>A estrutura intelectual do campo e as frentes emergentes de pesquisa foram caracterizadas por meio de um mapa temático, que classifica os domínios de pesquisa em quatro quadrantes de acordo com a densidade (grau de desenvolvimento) e a centralidade (relevância dentro da rede de pesquisa).</p>
					<p>O quadrante superior direito compreende os temas motores, caracterizados por elevada centralidade e elevada densidade, representando os tópicos mais influentes e bem desenvolvidos do campo. Os temas predominantes incluem <italic>cardiovascular magnetic resonance, artificial intelligence, magnetic resonance imaging, outcomes</italic> e <italic>position statement</italic> (<xref ref-type="fig" rid="f08002">Figura 7</xref>).</p>
					<p>
						<fig id="f08002">
							<label>Figura 7</label>
							<caption>
								<title>– Mapa temático das tendências de pesquisa e dos hotspots emergentes.</title>
							</caption>
							<graphic xlink:href="2764-3107-abchf-6-e20260016-gf08-pt.tif"/>
						</fig>
					</p>
					<p>O quadrante inferior direito contém os temas básicos, que são altamente relevantes para o campo, mas apresentam menor grau de desenvolvimento interno. Esse quadrante inclui <italic>deep learning, magnetic resonance imaging</italic> e <italic>strain</italic>. Na interface com os temas motores encontram-se agrupamentos relacionados a <italic>myocarditis, COVID-19, SARS-CoV-2, disease</italic> e <italic>fulminant myocarditis</italic>.</p>
					<p>O quadrante superior esquerdo contém os temas de nicho, que são altamente desenvolvidos, mas permanecem periféricos à rede geral de pesquisa. Entre eles estão <italic>endomyocardial biopsy</italic> e <italic>inflammatory cardiomyopathy</italic>. O quadrante inferior esquerdo compreende temas emergentes ou em declínio, cuja interpretação depende de sua evolução temporal. Essa região é dominada por termos amplos orientados à predição, incluindo <italic>classification, prediction, diagnosis</italic> e <italic>machine learning</italic>.</p>
					<p>As análises temporais baseadas na frequência das palavras-chave e nos gráficos de tendências temáticas identificaram um claro ponto de inflexão em 2021, correspondente ao acentuado aumento da produção científica, impulsionado pelo rápido crescimento dos termos <italic>myocarditis, diagnosis</italic> e <italic>artificial intelligence</italic>. A partir de 2023, <italic>deep learning</italic> e <italic>cardiovascular magnetic resonance</italic> tornaram-se progressivamente mais proeminentes. Entre todas as palavras-chave, <italic>machine learning</italic> apresentou o período mais longo de uso sustentado ao longo do intervalo estudado.</p>
				</sec>
			</sec>
			<sec sec-type="discussion">
				<title>Discussão</title>
				<p>Esta análise bibliométrica indica que a interseção entre inteligência artificial e miocardite representa um campo de pesquisa emergente e ainda relativamente imaturo. A produção científica aumentou acentuadamente após 2021, coincidindo com a pandemia de covid-19, que provavelmente atuou como um catalisador global para as pesquisas nessa área. O crescente interesse clínico na lesão miocárdica induzida por vírus, juntamente com a disponibilidade de grandes conjuntos de dados hospitalares gerados durante a pandemia, pode ter acelerado a adoção de algoritmos de aprendizado de máquina e aprendizado profundo para estratificação de risco e apoio diagnóstico em pacientes com suspeita de miocardite.<sup><xref ref-type="bibr" rid="B11">11</xref>,<xref ref-type="bibr" rid="B18">18</xref></sup></p>
				<p>O recente declínio no número médio de citações por artigo não deve ser interpretado como evidência de redução da relevância científica. Uma explicação mais plausível é a latência de citações, uma limitação inerente às análises bibliométricas.<sup><xref ref-type="bibr" rid="B19">19</xref></sup> Artigos publicados nos últimos 24 meses tiveram tempo limitado para acumular citações. Assim, o pico de citações observado em 2021 provavelmente reflete a base metodológica sobre a qual as pesquisas subsequentes nesse campo foram desenvolvidas.</p>
				<p>Apesar disso, a taxa atual de crescimento permanece modesta quando comparada ao campo mais amplo da cardiologia.<sup><xref ref-type="bibr" rid="B20">20</xref></sup> Esse achado sugere que as aplicações da inteligência artificial na miocardite ainda estão passando por um processo de consolidação metodológica. Espera-se que a expansão contínua da inteligência artificial na cardiologia ocorra nas próximas décadas, particularmente na imagem cardiovascular, área na qual se prevê que as abordagens baseadas em inteligência artificial desempenhem um papel cada vez mais proeminente.<sup><xref ref-type="bibr" rid="B3">3</xref>,<xref ref-type="bibr" rid="B21">21</xref></sup></p>
				<p>A análise das contribuições em nível de país revelou estratégias de pesquisa distintas entre as principais nações produtoras de conhecimento científico. A China respondeu pelo maior número de publicações, sugerindo uma forte ênfase nacional no desenvolvimento e na implementação de tecnologias diagnósticas baseadas em inteligência artificial. Em contraste, a rede de pesquisa europeia, liderada pelo Reino Unido e pela Alemanha, produziu menos publicações, mas alcançou impacto de citações substancialmente maior, com a maioria dos estudos mais citados originando-se desses países. Os Estados Unidos apresentaram um perfil de pesquisa predominantemente nacional, possivelmente refletindo maior dependência de conjuntos de dados institucionais e privados do que de consórcios internacionais de colaboração. Um achado notável foi a completa ausência de publicações provenientes da América Latina, do Caribe e da África. Esse desequilíbrio geográfico suscita importantes preocupações quanto à validade externa e à generalização dos modelos de inteligência artificial para miocardite, uma vez que os conjuntos de dados de treinamento e validação podem representar inadequadamente essas populações, aumentando, assim, o risco de viés algorítmico em sistemas de apoio à decisão clínica.<sup><xref ref-type="bibr" rid="B22">22</xref></sup></p>
				<p>A estrutura de autoria do campo também permanece altamente fragmentada. De acordo com a Lei de Lotka, 98,4% dos autores contribuíram com apenas uma única publicação sobre o tema, indicando um panorama de pesquisa caracterizado por contribuições isoladas, em vez de uma atividade contínua por parte de grupos de pesquisa dedicados.<sup><xref ref-type="bibr" rid="B9">9</xref></sup> O exame da formação acadêmica dos autores com maior impacto local destaca outra característica definidora do campo: sua acentuada interdisciplinaridade e o papel de liderança atualmente desempenhado por pesquisadores das áreas de engenharia, ciência da computação e outras disciplinas quantitativas no avanço da inteligência artificial aplicada à medicina.<sup><xref ref-type="bibr" rid="B23">23</xref></sup> Juan Manuel Górriz, que alcançou o maior índice M (1,0), possui formação em Física e Engenharia Eletrônica e é especialista em processamento estatístico de sinais para aplicações biomédicas.<sup><xref ref-type="bibr" rid="B24">24</xref></sup> Roohallah Alizadehsani, que liderou o campo tanto em número de publicações quanto em índice H, possui doutorado em Engenharia da Computação e concentra suas pesquisas no desenvolvimento de algoritmos de aprendizado de máquina para doenças cardiovasculares.<sup><xref ref-type="bibr" rid="B25">25</xref></sup></p>
				<p>Esse perfil sugere que o desenvolvimento metodológico e tecnológico dos modelos de inteligência artificial está atualmente avançando mais rapidamente do que sua validação clínica. O papel de destaque desempenhado por engenheiros e cientistas de dados indica que a pesquisa tem se concentrado principalmente no desenvolvimento, treinamento e otimização de modelos computacionais capazes de processar conjuntos de dados complexos, particularmente dados de imagem cardiovascular e outros dados clínicos de alta dimensionalidade.<sup><xref ref-type="bibr" rid="B26">26</xref></sup></p>
				<p>Um padrão semelhante é observado nos principais periódicos do campo. A disseminação científica está concentrada em periódicos voltados à inovação em imagem cardiovascular e tecnologias diagnósticas, refletindo a dependência dos atuais modelos de inteligência artificial de biomarcadores derivados de exames de imagem para a detecção da miocardite. A translação dessas tecnologias para a prática clínica rotineira, entretanto, provavelmente exigirá maior ênfase na pesquisa translacional.<sup><xref ref-type="bibr" rid="B3">3</xref></sup> A superação da lacuna entre o desenvolvimento computacional e a implementação à beira do leito dependerá de equipes multidisciplinares, nas quais a expertise em desenvolvimento de algoritmos seja integrada à interpretação fisiopatológica e à rigorosa curadoria de dados realizada por cardiologistas e pesquisadores clínicos.</p>
				<p>Essa evolução do paradigma diagnóstico também se reflete no mapa temático e nas redes de coocorrência de palavras-chave. Embora a biópsia endomiocárdica permaneça como o tradicional padrão-ouro diagnóstico, ela está posicionada em um agrupamento temático de nicho, enquanto as técnicas de imagem cardiovascular dominam os temas motores. As pesquisas atuais estão cada vez mais voltadas ao refinamento de algoritmos de aprendizado profundo e aprendizado de máquina para identificar padrões sutis em exames de imagem e melhorar o desempenho de abordagens diagnósticas não invasivas.<sup><xref ref-type="bibr" rid="B3">3</xref>,<xref ref-type="bibr" rid="B13">13</xref>,<xref ref-type="bibr" rid="B21">21</xref></sup></p>
				<p>Além de melhorar a acurácia diagnóstica, estudos recentes têm explorado o uso de inteligência artificial generativa e de técnicas de aumento de dados para enfrentar a limitada disponibilidade de dados referentes a apresentações clínicas raras ou subdiagnosticadas, incluindo miocardite fulminante e miocardite associada a inibidores de <italic>checkpoint</italic> imunológico.<sup><xref ref-type="bibr" rid="B27">27</xref>,<xref ref-type="bibr" rid="B28">28</xref></sup> Paralelamente, o destaque da palavra-chave <italic>outcomes</italic> entre os temas motores reflete o crescente interesse pela modelagem prognóstica e pela predição de eventos clínicos. Por fim, a coocorrência de termos relacionados à inteligência artificial com palavras-chave orientadas por diretrizes, como <italic>position statement</italic> e <italic>working group</italic>, sugere uma ênfase crescente no estabelecimento de estruturas padronizadas para a avaliação, validação e relato de aplicações de inteligência artificial em imagem cardiovascular.<sup><xref ref-type="bibr" rid="B29">29</xref></sup></p>
				<sec>
					<title>Limitações</title>
					<p>Diversas limitações deste estudo devem ser reconhecidas. Primeiramente, a extração dos dados foi restrita à Web of Science Core Collection; portanto, artigos relevantes indexados exclusivamente em bases de dados como Scopus, Embase ou PubMed podem não ter sido capturados. Ainda assim, a Web of Science foi selecionada por fornecer os metadados padronizados necessários para análises robustas de redes bibliométricas e de citações.<sup><xref ref-type="bibr" rid="B6">6</xref></sup></p>
					<p>Em segundo lugar, a estratégia de busca baseou-se em palavras-chave predefinidas. Consequentemente, estudos que descrevem técnicas computacionais específicas (p.ex., <italic>random forests</italic> ou SVMs), sem incluir explicitamente os termos mais amplos <italic>artificial intelligence</italic> ou <italic>machine learning</italic> no título, resumo ou palavras-chave, podem ter sido excluídos.</p>
					<p>Em terceiro lugar, as análises bibliométricas estão inerentemente sujeitas ao viés de latência de citações,<sup><xref ref-type="bibr" rid="B19">19</xref></sup> pelo qual artigos publicados recentemente dispõem de tempo insuficiente para acumular citações, independentemente de sua qualidade científica ou de seu impacto potencial. Essa limitação foi parcialmente abordada por meio da incorporação do índice M, que considera a natureza dependente do tempo da acumulação de citações.</p>
					<p>Por fim, o número relativamente pequeno de publicações incluídas (n = 47) pode limitar a generalização dos achados. Entretanto, essa característica também reflete o estágio inicial de desenvolvimento da pesquisa em inteligência artificial aplicada à miocardite, destacando o surgimento de um campo de pesquisa especializado e ainda em expansão.</p>
				</sec>
			</sec>
			<sec sec-type="conclusions">
				<title>Conclusão</title>
				<p>Esta análise bibliométrica fornece uma visão abrangente do panorama científico global na interseção entre inteligência artificial e miocardite. O campo permanece em estágio inicial de desenvolvimento, caracteriza-se por marcadas disparidades geográficas e é atualmente impulsionado predominantemente por pesquisadores das áreas de engenharia e ciência da computação, em vez de cardiologistas clínicos. Sua expansão recente coincidiu com a pandemia de covid-19, enquanto a ressonância magnética cardiovascular emergiu como a principal fonte de dados que sustenta os modelos diagnósticos baseados em inteligência artificial.</p>
				<p>A translação bem-sucedida dessas tecnologias para a prática clínica rotineira provavelmente dependerá da colaboração multidisciplinar e da validação rigorosa em populações diversas, particularmente em regiões que permanecem sub-representadas na literatura atual, como América Latina, Caribe e África. Embora este estudo tenha sido limitado a uma única base de dados bibliográfica e tenha incluído um número relativamente pequeno de publicações, esses achados fornecem uma base valiosa para futuras investigações e contribuem para uma melhor compreensão da evolução, do estado atual e das perspectivas futuras desse campo em rápido desenvolvimento.</p>
			</sec>
		</body>
		<back>
			<fn-group>
				<fn fn-type="other">
					<label>Vinculação Acadêmica:</label>
					<p> Não há vinculação deste estudo a programas de pós-graduação.</p>
				</fn>
				<fn fn-type="other">
					<label>Aprovação Ética e Consentimento Informado:</label>
					<p> Este artigo não contém estudos com humanos ou animais realizados por nenhum dos autores.</p>
				</fn>
				<fn fn-type="other">
					<label>Uso de Inteligência Artificial:</label>
					<p> Os autores não utilizaram ferramentas de inteligência artificial no desenvolvimento deste trabalho.</p>
				</fn>
				<fn fn-type="data-availability" specific-use="data-available-upon-request">
					<label>Disponibilidade de Dados:</label>
					<p> Todo o conjunto de dados que dá suporte aos resultados deste estudo está disponível mediante solicitação ao autor correspondente.</p>
				</fn>
				<fn fn-type="financial-disclosure">
					<label>Fontes de Financiamento:</label>
					<p> O presente estudo não teve fontes de financiamento externas.</p>
				</fn>
			</fn-group>
		</back>
	</sub-article>
</article>