<!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 xmlns:xlink="http://www.w3.org/1999/xlink" specific-use="sps-1.9" dtd-version="1.1" xml:lang="pt" article-type="research-article">
  <front>
    <journal-meta>
      <journal-id journal-id-type="publisher-id">abc</journal-id>
      <journal-title-group>
        <journal-title>Arquivos Brasileiros de Cardiologia</journal-title>
        <abbrev-journal-title abbrev-type="publisher">Arq. Bras. Cardiol.</abbrev-journal-title>
      </journal-title-group>
      <issn pub-type="ppub">0066-782X</issn>
      <issn pub-type="epub">1678-4170</issn>
      <publisher>
        <publisher-name>Sociedade Brasileira de Cardiologia - SBC</publisher-name>
        <publisher-loc>S&#227;o Paulo, SP, Brazil</publisher-loc>
      </publisher>
    </journal-meta>
    <article-meta>
      <article-id pub-id-type="publisher-id" specific-use="scielo-v3">yM8bQtbszJBMVkgrzMp88GL</article-id>
      <article-id pub-id-type="publisher-id" specific-use="scielo-v2">S0066-782X2012001300011</article-id>
      <article-id pub-id-type="publisher-id" specific-use="scielo-v1">S0066-782X(12)09900411</article-id>
      <article-id pub-id-type="doi">10.1590/S0066-782X2012005000089</article-id>
      <article-categories>
        <subj-group subj-group-type="heading">
          <subject>Artigos Originais</subject>
        </subj-group>
      </article-categories>
      <title-group>
        <article-title>Software livre e de c&#243;digo aberto para avalia&#231;&#227;o de imagens de angiotomografia de coron&#225;rias</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <name>
            <surname>Hadlich</surname>
            <given-names>Marcelo Souza</given-names>
          </name>
          <xref ref-type="aff" rid="aff01"/>
          <xref ref-type="aff" rid="aff02"/>
          <xref ref-type="aff" rid="aff04"/>
          <xref ref-type="aff" rid="aff05"/>
          <xref ref-type="aff" rid="aff06"/>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Oliveira</surname>
            <given-names>Gl&#225;ucia Maria Moraes</given-names>
          </name>
          <xref ref-type="aff" rid="aff01"/>
          <xref ref-type="aff" rid="aff02"/>
          <xref ref-type="aff" rid="aff04"/>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Feij&#243;o</surname>
            <given-names>Ra&#250;l A.</given-names>
          </name>
          <xref ref-type="aff" rid="aff03"/>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Azevedo</surname>
            <given-names>Clerio F.</given-names>
          </name>
          <xref ref-type="aff" rid="aff04"/>
          <xref ref-type="aff" rid="aff05"/>
          <xref ref-type="aff" rid="aff06"/>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Tura</surname>
            <given-names>Bernardo Rangel</given-names>
          </name>
          <xref ref-type="aff" rid="aff05"/>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Ziemer</surname>
            <given-names>Paulo Gustavo Portela</given-names>
          </name>
          <xref ref-type="aff" rid="aff03"/>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Blanco</surname>
            <given-names>Pablo Javier</given-names>
          </name>
          <xref ref-type="aff" rid="aff03"/>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Pina</surname>
            <given-names>Gustavo</given-names>
          </name>
          <xref ref-type="aff" rid="aff01"/>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Meira</surname>
            <given-names>M&#225;rcio</given-names>
          </name>
          <xref ref-type="aff" rid="aff01"/>
        </contrib>
        <contrib contrib-type="author">
          <name>
            <surname>Souza e Silva</surname>
            <given-names>Nelson Albuquerque de</given-names>
          </name>
          <xref ref-type="aff" rid="aff01"/>
          <xref ref-type="aff" rid="aff02"/>
        </contrib>
      </contrib-group>
      <aff id="aff01">
        <addr-line>
          <named-content content-type="city">Rio de Janeiro</named-content>
          <named-content content-type="state"/>
        </addr-line>
        <institution content-type="orgname">Universidade Federal do Rio de Janeiro</institution>
        <country country="BR">Brazil</country>
      </aff>
      <aff id="aff02">
        <addr-line>
          <named-content content-type="city">Rio de Janeiro</named-content>
          <named-content content-type="state">RJ</named-content>
        </addr-line>
        <institution content-type="orgname">Instituto do Cora&#231;&#227;o Edson Saad</institution>
        <country country="BR">Brasil</country>
      </aff>
      <aff id="aff05">
        <addr-line>
          <named-content content-type="city">Rio de Janeiro</named-content>
          <named-content content-type="state"/>
        </addr-line>
        <institution content-type="orgname">Instituto Nacional de Cardiologia</institution>
        <country country="BR">Brazil</country>
      </aff>
      <aff id="aff04">
        <addr-line>
          <named-content content-type="city">Rio de Janeiro</named-content>
          <named-content content-type="state">RJ</named-content>
        </addr-line>
        <institution content-type="orgname">Rede Labs Dor</institution>
        <country country="BR">Brasil</country>
      </aff>
      <aff id="aff03">
        <addr-line>
          <named-content content-type="city">Rio de Janeiro</named-content>
          <named-content content-type="state"/>
        </addr-line>
        <institution content-type="orgname">Laborat&#243;rio Nacional de Computa&#231;&#227;o Cient&#237;fica</institution>
        <country country="BR">Brazil</country>
      </aff>
      <aff id="aff06">
        <addr-line>
          <named-content content-type="city">Rio de Janeiro</named-content>
          <named-content content-type="state">Rio de Janeiro</named-content>
        </addr-line>
        <institution content-type="orgname">Instituto Dor Pesquisa e Ensino</institution>
        <country country="BR">Brazil</country>
      </aff>
      <pub-date date-type="collection" publication-format="electronic">
        <month>10</month>
        <year>2012</year>
      </pub-date>
      <volume>99</volume>
      <issue>4</issue>
      <fpage>944</fpage>
      <lpage>951</lpage>
      <history>
        <date date-type="received">
          <day>04</day>
          <month>11</month>
          <year>2011</year>
        </date>
        <date date-type="accepted">
          <day>13</day>
          <month>06</month>
          <year>2012</year>
        </date>
        <date date-type="rev-recd">
          <day>04</day>
          <month>11</month>
          <year>2011</year>
        </date>
      </history>
      <permissions>
        <license license-type="open-access" xlink:href="http://creativecommons.org/licenses/by-nc/4.0/" xml:lang="en">
          <license-p>This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.</license-p>
        </license>
      </permissions>
      <abstract>
        <p>FUNDAMENTO: A estandardiza&#231;&#227;o do padr&#227;o de imagens utilizada dentro da medicina foi realizada em 1993 por meio do padr&#227;o DICOM (Digital Imaging and Communications in Medicine). Diversos exames utilizam esse padr&#227;o e cada vez mais s&#227;o necess&#225;rios softwares capazes de manipular esse tipo de imagem, por&#233;m esses softwares geralmente n&#227;o t&#234;m o formato livre e de c&#243;digo aberto, e isso dificulta o seu ajuste para os mais diferentes interesses. OBJETIVO: Desenvolver e validar um software livre e de c&#243;digo aberto capaz de manipular imagens DICOM de exames de angiotomografia de coron&#225;rias. M&#201;TODOS: Desenvolvemos e testamos o software intitulado ImageLab na avalia&#231;&#227;o de 100 exames selecionados de forma rand&#244;mica por meio de um banco de dados. Foram realizadas 600 an&#225;lises divididas por dois observadores utilizando o ImageLab e um outro software comercializado junto a aparelhos de tomografia computadorizada Philips Brilliance, na avalia&#231;&#227;o da presen&#231;a de les&#245;es e placas coronarianas nos territ&#243;rios do Tronco da Coron&#225;ria Esquerda (TCE) e na Art&#233;ria Descendente Anterior (ADA). Para avaliar as concord&#226;ncias intraobservador, interobservadores e intersoftware, utilizamos concord&#226;ncia simples e estat&#237;stica Kappa. RESULTADOS: As concord&#226;ncias observadas entre os softwares foram em geral classificadas como substancial ou quase perfeitas na maioria das compara&#231;&#245;es. CONCLUS&#195;O: O software ImageLab concordou com o software Philips na avalia&#231;&#227;o de exames de angiotomografia de coron&#225;rias especialmente em pacientes sem les&#245;es, com les&#245;es inferiores a 50% no TCE e inferiores a 70% na ADA. A concord&#226;ncia para les&#227;o &gt;70% na ADA foi menor, por&#233;m isso tamb&#233;m &#233; observado quando se utiliza o padr&#227;o de refer&#234;ncia anat&#244;mico.</p>
      </abstract>
      <kwd-group xml:lang="pt" kwd-group-type="author-generated">
        <kwd>Diagn&#243;stico por imagem</kwd>
        <kwd>tomografia computadorizada</kwd>
        <kwd>vasos coron&#225;rios</kwd>
        <kwd>valida&#231;&#227;o de programas de computador</kwd>
      </kwd-group>
      <pub-date publication-format="electronic" date-type="pub">
        <day>02</day>
        <month>10</month>
        <year>2012</year>
      </pub-date>
    </article-meta>
  </front>
  <body>
    <p>
      <bold>Software livre e de c&#243;digo aberto para avalia&#231;&#227;o de imagens de angiotomografia de coron&#225;rias</bold>
    </p>
    <p>
      <bold>Marcelo Souza Hadlich<sup>I,II,IV,V,VI</sup>; Gl&#225;ucia Maria Moraes Oliveira<sup>I,II,IV</sup>; Ra&#250;l A. Feij&#243;o<sup>III</sup>; Clerio F. Azevedo<sup>IV,V,VI</sup>; Bernardo Rangel Tura<sup>V</sup>; Paulo Gustavo Portela Ziemer<sup>III</sup>; Pablo Javier Blanco<sup>III</sup>; Gustavo Pina<sup>I</sup>; M&#225;rcio Meira<sup>I</sup>, Nelson Albuquerque de Souza e Silva<sup>I,II</sup></bold>
    </p>
    <p><sup>I</sup>Universidade Federal do Rio de Janeiro (UFRJ). Rio de Janeiro, RJ - Brasil </p>
    <p><sup>II</sup>Instituto do Cora&#231;&#227;o Edson Saad. Rio de Janeiro, RJ - Brasil </p>
    <p><sup>III</sup>Laborat&#243;rio Nacional de Computa&#231;&#227;o Cient&#237;fica (LNCC). Rio de Janeiro, RJ - Brasil </p>
    <p><sup>IV</sup>Rede Labs Dor. Rio de Janeiro, RJ - Brasil </p>
    <p><sup>V</sup>Instituto Nacional de Cardiologia. Rio de Janeiro, RJ - Brasil </p>
    <p><sup>VI</sup>Instituto Dor de Pesquisa e Ensino. Rio de Janeiro, RJ - Brasil</p>
    <p>
      <xref ref-type="corresp" rid="end">Correspond&#234;ncia</xref>
    </p>
    <p>
      <bold>RESUMO</bold>
    </p>
    <p><bold>FUNDAMENTO: </bold> A estandardiza&#231;&#227;o do padr&#227;o de imagens utilizada dentro da medicina foi realizada em 1993 por meio do padr&#227;o DICOM (Digital Imaging and Communications in Medicine). Diversos exames utilizam esse padr&#227;o e cada vez mais s&#227;o necess&#225;rios softwares capazes de manipular esse tipo de imagem, por&#233;m esses softwares geralmente n&#227;o t&#234;m o formato livre e de c&#243;digo aberto, e isso dificulta o seu ajuste para os mais diferentes interesses. </p>
    <p><bold>OBJETIVO:</bold> Desenvolver e validar um software livre e de c&#243;digo aberto capaz de manipular imagens DICOM de exames de angiotomografia de coron&#225;rias. </p>
    <p><bold>M&#201;TODOS:</bold> Desenvolvemos e testamos o software intitulado ImageLab na avalia&#231;&#227;o de 100 exames selecionados de forma rand&#244;mica por meio de um banco de dados. Foram realizadas 600 an&#225;lises divididas por dois observadores utilizando o ImageLab e um outro software comercializado junto a aparelhos de tomografia computadorizada Philips Brilliance, na avalia&#231;&#227;o da presen&#231;a de les&#245;es e placas coronarianas nos territ&#243;rios do Tronco da Coron&#225;ria Esquerda (TCE) e na Art&#233;ria Descendente Anterior (ADA). Para avaliar as concord&#226;ncias intraobservador, interobservadores e intersoftware, utilizamos concord&#226;ncia simples e estat&#237;stica Kappa. </p>
    <p><bold>RESULTADOS:</bold> As concord&#226;ncias observadas entre os softwares foram em geral classificadas como substancial ou quase perfeitas na maioria das compara&#231;&#245;es. </p>
    <p><bold>CONCLUS&#195;O: </bold>O software ImageLab concordou com o software Philips na avalia&#231;&#227;o de exames de angiotomografia de coron&#225;rias especialmente em pacientes sem les&#245;es, com les&#245;es inferiores a 50% no TCE e inferiores a 70% na ADA. A concord&#226;ncia para les&#227;o &gt;70% na ADA foi menor, por&#233;m isso tamb&#233;m &#233; observado quando se utiliza o padr&#227;o de refer&#234;ncia anat&#244;mico. </p>
    <p><bold>Palavras-chave:</bold> Diagn&#243;stico por imagem; tomografia computadorizada; vasos coron&#225;rios; valida&#231;&#227;o de programas de computador.</p>
    <p>
      <bold>Introdu&#231;&#227;o</bold>
    </p>
    <p>Na &#225;rea m&#233;dica utiliza-se um padr&#227;o espec&#237;fico de imagem chamado DICOM (Digital Imaging and Communications in Medicine). Esse padr&#227;o foi desenvolvido para facilitar a comunica&#231;&#227;o entre os softwares e hardwares relacionados com esse processo.</p>
    <p>O formato DICOM, padronizado<sup>1</sup> em 1993 junto ao congresso do RSNA (Radiological Society of North America), teve como objetivo a normatiza&#231;&#227;o das regras com que as informa&#231;&#245;es m&#233;dicas s&#227;o transmitidas e armazenadas<sup>2-4</sup>.</p>
    <p>A visualiza&#231;&#227;o e o processamento das imagens m&#233;dicas s&#227;o realizados por meio de softwares espec&#237;ficos. Grande parte desses recursos n&#227;o est&#225; dispon&#237;vel de forma gratuita ou &#233; vendida junto aos equipamentos, por&#233;m n&#227;o &#233; comum encontrar softwares livres e de c&#243;digo aberto para esse formato de imagens. O dom&#237;nio p&#250;blico da tecnologia de um software m&#233;dico poderia diminuir custos, homogeneizar o uso multic&#234;ntrico, permitir desenvolvimento cont&#237;nuo e facilitar o desenvolvimento de diversas linhas de pesquisa, e alguns acreditam que esse formato poderia contribuir para a diminui&#231;&#227;o da diferen&#231;a social entre pa&#237;ses<sup>5</sup>.</p>
    <p>O Instituto Nacional de Ci&#234;ncia e Tecnologia em Medicina Assistida por Computa&#231;&#227;o Cient&#237;fica (INCT-MACC) envolve 33 institui&#231;&#245;es nacionais de 11 Estados, totalizando 128 pesquisadores. Uma das &#225;reas de pesquisa do INCT-MACC &#233; a de processamento de imagens m&#233;dicas, que permitir&#225; a aquisi&#231;&#227;o de informa&#231;&#245;es para aperfei&#231;oar a modelagem computacional do sistema cardiovascular humano, tamb&#233;m em desenvolvimento pelo INCT-MACC. Com essa finalidade foi elaborado um software livre e de c&#243;digo aberto denominado ImageLab. Esse software teria entre caracter&#237;sticas ser amig&#225;vel com o usu&#225;rio necessitando de melhor avalia&#231;&#227;o.</p>
    <p>O objetivo do estudo foi desenvolver e avaliar a confiabilidade, a concord&#226;ncia e a reprodutibilidade do novo software em rela&#231;&#227;o a outro em utiliza&#231;&#227;o corrente em equipamentos de angiotomografia de coron&#225;rias.</p>
    <p>
      <bold>M&#233;todos</bold>
    </p>
    <p>Em trabalho conjunto da Universidade Federal do Rio de Janeiro (UFRJ) com o Laborat&#243;rio Nacional de Computa&#231;&#227;o Cient&#237;fica (LNCC), foi desenvolvido um software para an&#225;lise de imagens DICOM. Esse software livre de c&#243;digo aberto chama-se ImageLab e possui ferramentas criadas para an&#225;lise e processamento de imagens de angiotomografia de coron&#225;rias.</p>
    <p>De uma popula&#231;&#227;o de 6.216 pacientes que realizaram exame de angiotomografia de coron&#225;rias entre maio de 2005 e dezembro de 2010, aplicamos o crit&#233;rio de inclus&#227;o, ou seja, pacientes com dados completos registrados em banco de dados criado em 2008, e selecionamos 2.895 pacientes. Posteriormente ap&#243;s aplicarmos os crit&#233;rios de exclus&#227;o, presen&#231;a de <italic>stents</italic> no tronco da coron&#225;ria esquerda ou em Art&#233;ria Descendente Anterior (ADA), clipe cir&#250;rgico, decorrente de cirurgia de revasculariza&#231;&#227;o mioc&#225;rdica e pacientes que n&#227;o possu&#237;am imagens armazenadas no servidos, foram selecionados 534 casos (<xref ref-type="fig" rid="aop08612fig01">Figura 1</xref>).</p>
    <p content-type="asset">
      <fig id="aop08612fig01">
        <graphic xlink:href="1678-4170-abc-99-04-0944-00089-gf01.jpg"/></fig>
    </p>
    <p>Pelo <italic>site</italic> randon.org realizamos sele&#231;&#227;o rand&#244;mica retirando 120 dos 534 pacientes, e utilizamos os 100 primeiros casos. Todos os pacientes tiveram seus dados identificadores substitu&#237;dos por n&#250;meros consecutivos, por pessoas que n&#227;o participaram da an&#225;lise das imagens tornando-a cega.</p>
    <p>Foram utilizados os dados brutos de imagens de angiotomografia de coron&#225;rias, obtidos por meio de tom&#243;grafos dispon&#237;veis no servi&#231;o (GE lightspeed VTC 64 e 32, Philips Brilliance 64, 40 e 16). Todas as imagens seguiram protocolo padr&#227;o de aquisi&#231;&#227;o utilizando t&#233;cnicas semelhantes.</p>
    <p>Todos os 100 exames foram analisados por dois softwares. O software Philips Brilliance, comprado junto com o pacote cardiovascular (dispon&#237;vel em Worsktation pr&#243;pria para an&#225;lises de imagens do tipo DICOM). E o software ImageLab, desenvolvido nesse projeto, instalado na mesma Worsktation (visando neutralizar a poss&#237;vel varia&#231;&#227;o de capacidade de processamento caso fossem instalados em computadores com capacidade diferente).</p>
    <p>Escolhemos o tronco da coron&#225;ria esquerda e a ADA pela import&#226;ncia cl&#237;nica e para viabilizar a an&#225;lise, visto que a utiliza&#231;&#227;o dos 16 segmentos na sua totalidade resultaria em um grande n&#250;mero de vari&#225;veis. Seguimos a nomenclatura proposta pelo AHA, referente aos segmentos 5, 6, 7 e 8<sup>6</sup>. A avalia&#231;&#227;o da presen&#231;a de les&#245;es coronarianas maiores ou menores que 50% foi utilizada para o segmento 5 (tronco da coron&#225;ria esquerda) e para os demais segmentos (art&#233;ria descendente anterior) utilizamos cortes de 70%. De forma adicional tamb&#233;m foram registradas as vari&#225;veis: presen&#231;a de placas ateroscler&#243;ticas tipo calcificadas, n&#227;o calcificadas e parcialmente calcificadas (em cada segmento); qualidade subjetiva das imagens; tempo de an&#225;lise das imagens (registrado pelo observador 1).</p>
    <p>A an&#225;lise das imagens foi feita pela visualiza&#231;&#227;o de planos ortogonais selecionados a partir de bloco de imagem no espa&#231;o tridimensional, come&#231;ando pela an&#225;lise em plano transversal do t&#243;rax no sentido cabe&#231;a para os p&#233;s, de modo a visualizar a art&#233;ria aorta, a origem do tronco da coron&#225;ria esquerda e a ADA. As avalia&#231;&#245;es referentes ao grau de estenose seguem o padr&#227;o de an&#225;lise onde a falha de enchimento junto ao vaso contrastado define a presen&#231;a de les&#227;o. A an&#225;lise do tipo de placa respons&#225;vel pelas les&#245;es foi realizada de forma subjetiva. A caracteriza&#231;&#227;o das placas foi realizada de acordo com sua morfologia e intensidade de sinal, seguindo a escala tradicional adotada em exames de TC, a escala de Hounsfield (HU)<sup>7</sup>. As placas foram classificadas como: calcificadas (tecido adjacente ao vaso com mais intensidade de sinal que o vaso contrastado - Sinal &gt;130 HU); n&#227;o calcificadas (tecido adjacente ao vaso com menos sinal que o vaso contrastado - Sinal &lt;-50 HU); e parcialmente calcificada (conte&#250;do heterog&#234;neo).</p>
    <p>Os dois softwares apresentam ferramenta para a localiza&#231;&#227;o e an&#225;lise de qualquer ponto no espa&#231;o tridimensional. Essa ferramenta foi utilizada para a avalia&#231;&#227;o dos vasos, no seu plano axial (visualiza&#231;&#227;o de dentro dos vasos) em toda a sua extens&#227;o, e quando existir a presen&#231;a de placas e les&#245;es coronarianas, todos os planos ortogonais poder&#227;o ser utilizados de forma simult&#226;nea, na topografia exata em que esses achados se encontrem.</p>
    <p>
      <bold>An&#225;lise das Imagens</bold>
    </p>
    <p>Dois observadores com experi&#234;ncia superior a cinco anos, equivalente ao n&#237;vel 3 de compet&#234;ncia cl&#237;nica<sup>8</sup>, utilizaram software comercial da Philips e o ImageLab, instalados no mesmo computador, para analisar o segmento da ADA de acordo com as vari&#225;veis descritas na <xref ref-type="table" rid="aop08612tab01">Tabela 1</xref>. Para evitar vi&#233;s de aferi&#231;&#227;o, as an&#225;lises entre os softwares ocorreram com intervalos superiores a 15 dias.</p>
    <p content-type="asset">
      <table-wrap id="aop08612tab01">
        <graphic xlink:href="1678-4170-abc-99-04-0944-00089-gt01.jpg"/></table-wrap>
    </p>
    <p>Um total de 600 an&#225;lises foram realizadas para avaliar as concord&#226;ncias intraobservador, interobservador e intersoftware.</p>
    <p>
      <bold>An&#225;lise Estat&#237;stica</bold>
    </p>
    <p>An&#225;lise descritiva como vari&#225;veis num&#233;ricas, na forma de m&#233;dia +/- desvio padr&#227;o. Vari&#225;veis categ&#243;ricas como n&#250;mero (n) e percentagens (%). Estat&#237;stica Kappa para calcular a reprodutibilidade interobservador e intraobservador, todos com intervalo de confian&#231;a de 95% e concord&#226;ncia simples para medir os casos com n&#250;mero pequeno de discord&#226;ncia. O programa utilizado foi o R para Linux. Utilizamos a <xref ref-type="table" rid="aop08612tab01">Tabela 1</xref> para avaliar o grau de acordo entre as concord&#226;ncias<sup>9</sup>.</p>
    <p>
      <bold>Resultados</bold>
    </p>
    <p>Ap&#243;s a sele&#231;&#227;o rand&#244;mica dos 120 casos, 7 casos com numera&#231;&#227;o inferior a 100 foram exclu&#237;dos por falta da fase contrastada do exame (exames da modalidade escore de c&#225;lcio sem a angiotomografia de coron&#225;rias). Um total de 100 exames com numera&#231;&#227;o de 1 a 107 foram inclu&#237;dos para as 600 an&#225;lises divididas pelos observadores e softwares. (Todos os resultados das concord&#226;ncias obtidos com o programa R for Linux podem ser acessados pelo link <ext-link ext-link-type="uri" xlink:href="http://cl.ly/3c193E0J1o1M0f360u1d">http://cl.ly/3c193E0J1o1M0f360u1d</ext-link>)</p>
    <p>O resultado da vers&#227;o do software ImageLab utilizada neste projeto tornou-se adequado apresentando visual e usabilidade convenientes (<xref ref-type="fig" rid="aop08612fig02">Figura 2</xref>).</p>
    <p content-type="asset">
      <fig id="aop08612fig02">
        <graphic xlink:href="1678-4170-abc-99-04-0944-00089-gf02.jpg"/></fig>
    </p>
    <p>Todos os 100 pacientes analisados realizaram os exames entre julho de 2009 e novembro de 2010, a maioria do sexo masculino (65) e com a m&#233;dia de idade de 58 anos (23 a 85). O &#237;ndice de massa corporal m&#233;dio foi de 27 Kg/m<sup>2</sup>, classificando a m&#233;dia da popula&#231;&#227;o com sobrepeso (25% classificados como obesos). Apenas 15% da popula&#231;&#227;o n&#227;o apresentavam nenhum fator de risco, e 35% eram assintom&#225;ticos (<xref ref-type="table" rid="aop08612tab02">Tabela. 2</xref>).</p>
    <p content-type="asset">
      <table-wrap id="aop08612tab02">
        <graphic xlink:href="1678-4170-abc-99-04-0944-00089-gt02.jpg"/></table-wrap>
    </p>
    <p>O sintoma mais frequente observado foi a presen&#231;a de dor at&#237;pica para Doen&#231;a Arterial Coronariana (DAC), e a maioria das indica&#231;&#245;es dos exames ocorreu para avalia&#231;&#227;o de algum sintoma. Cabe ressaltar que 36% dos pacientes foram encaminhados em raz&#227;o da presen&#231;a de teste funcional alterado, e 16%, para estratifica&#231;&#227;o de risco (avalia&#231;&#227;o de rotina sem rela&#231;&#227;o com sintomas nem exames alterados); apenas 2%, para excluir o diagn&#243;stico de doen&#231;a aterotromb&#243;tica como etiologia de uma cardiomiopatia; e 2%, em avalia&#231;&#227;o pr&#233;-operat&#243;ria (avalia&#231;&#227;o do risco cir&#250;rgico cardiovascular).</p>
    <p>O observador 1 utilizou um dispositivo eletr&#244;nico em formato de prancheta (tablet da Apple<sup>&#174;</sup> - Ipad 64 Gb Wifi) para registrar as an&#225;lises pelo software Bento<sup>&#174;</sup> for Ipad. Isso permitiu cadastrar o tempo das an&#225;lises de forma simples (<xref ref-type="table" rid="aop08612tab03">Tabela 3</xref>).</p>
    <p content-type="asset">
      <table-wrap id="aop08612tab03">
        <graphic xlink:href="1678-4170-abc-99-04-0944-00089-gt03.jpg"/></table-wrap>
    </p>
    <p>O tempo total das an&#225;lises realizadas pelo observador 1 foi de aproximadamente 6,8 horas, distribu&#237;das em 14 dias. O tempo m&#233;dio gasto com cada an&#225;lise foi de 2 minutos e 4 segundos (21 a 612 segundos). O tempo m&#233;dio utilizado com o software ImageLab para a an&#225;lise de todos os casos foi de 226,2 minutos, enquanto com o software Philips Brilliance foi de 180,9 (20% a menos). A segunda an&#225;lise por ambos os softwares foi mais r&#225;pida, com a diminui&#231;&#227;o do tempo total em 11,9 minutos, e 10,7 minutos nos softwares ImageLab e Philips Brilliance, respectivamente.</p>
    <p>Registramos a qualidade subjetiva das imagens por meio de escore de 1 a 3; 1 baixa qualidade, 2 qualidade intermedi&#225;ria, e 3 alta qualidade. Vinte e quatro por cento das imagens foram classificadas como baixa qualidade; 64,2%, como intermedi&#225;ria; e 11,8%, como alta qualidade (76% com qualidade intermedi&#225;ria ou alta).</p>
    <p>Para avaliar as concord&#226;ncias intraobservador e interobservadores, utilizamos a concord&#226;ncia simples (soma dos achados concordantes devido ao n&#250;mero total de casos), na presen&#231;a de poucos ou nenhum caso discordante (distribui&#231;&#227;o heterog&#234;nea na tabela 2 x 2 com poucos ou nenhum caso em uma das casas), e estat&#237;stica Kappa utilizando (<xref ref-type="table" rid="aop08612tab01">Tabela 1</xref>) como refer&#234;ncia do grau de acordo entre os resultados<sup>8</sup>.</p>
    <p>A m&#233;dia das observa&#231;&#245;es das 600 an&#225;lises (<xref ref-type="table" rid="aop08612tab03">Tabela 3</xref>) demonstrou aus&#234;ncia de les&#227;o no tronco da coron&#225;ria esquerda em 82%, e em ADA em 49,3%. Menos de 1% com les&#227;o &gt;50% no tronco da coron&#225;ria esquerda e 9% com les&#227;o &gt;70% em algum segmento da ADA. O tipo de placa mais observado foi a calcificada. </p>
    <p>Foi poss&#237;vel realizar a an&#225;lise de concord&#226;ncia por meio do Kappa, para todos os tipos de compara&#231;&#227;o (intraobservador, interobservador e intersoftware), nas vari&#225;veis: aus&#234;ncia de les&#227;o e les&#227;o &lt;50% do tronco da coron&#225;ria esquerda; aus&#234;ncia de les&#227;o da ADA e les&#227;o &gt;70% no ter&#231;o m&#233;dio dessa art&#233;ria (<xref ref-type="table" rid="aop08612tab04">Tabela 4</xref>). Na avalia&#231;&#227;o das placas o Kappa pode ser realizado apenas com a vari&#225;vel placa calcificada no tronco da coron&#225;ria esquerda (<xref ref-type="table" rid="aop08612tab05">Tabela 5</xref>). As demais concord&#226;ncias foram realizadas mediante a medida da concord&#226;ncia simples.</p>
    <p content-type="asset">
      <table-wrap id="aop08612tab04">
        <graphic xlink:href="1678-4170-abc-99-04-0944-00089-gt04.jpg"/></table-wrap>
    </p>
    <p content-type="asset">
      <table-wrap id="aop08612tab05">
        <graphic xlink:href="1678-4170-abc-99-04-0944-00089-gt05.jpg"/></table-wrap>
    </p>
    <p>Com o objetivo de avaliar as concord&#226;ncias intraobservador comparamos as an&#225;lises realizadas pelo observador 1 com o software ImageLab no tempo 1 e ImageLab no tempo 2, assim como o software Philips Brilliance no tempo 1 e Philips Brilliance no tempo 2</p>
    <p>Todas as avalia&#231;&#245;es medidas pelo Kappa ou pela concord&#226;ncia simples no tronco da coron&#225;ria esquerda foram superior a 60. Apenas a vari&#225;vel les&#227;o &gt;70% no ter&#231;o m&#233;dio da art&#233;ria descendente, na avalia&#231;&#227;o interobservador pelo software da Philips Brilliance, apresentou concord&#226;ncia moderada (49,7 - 13,8 a 85,6), todas as demais concord&#226;ncias observadas na ADA anterior foram superior a 60<sup>9</sup>. </p>
    <p>A avalia&#231;&#227;o de aus&#234;ncia de les&#245;es e les&#227;o &lt;50% no tronco da coron&#225;ria esquerda, assim como a avalia&#231;&#227;o de les&#227;o &gt;70% no ter&#231;o m&#233;dio da ADA demonstraram maior concord&#226;ncia intraobservador pelo softwares da Philips Brilliance, as demais mantiveram-se semelhante.</p>
    <p>A an&#225;lise intersoftware demonstrou concord&#226;ncia superior a 80 para todas as medidas realizadas pelo Kappa, exceto na avalia&#231;&#227;o de les&#227;o &gt;70% no ter&#231;o m&#233;dio de ADA (79,6 - 63,6 a 95,7), enquanto todas as concord&#226;ncias simples intersoftware foram superiores a 80.</p>
    <p>Quando analisadas as observa&#231;&#245;es mais frequentes e com maior signific&#226;ncia cl&#237;nica (<xref ref-type="table" rid="aop08612tab06">Tabela 6</xref>), foi poss&#237;vel utilizar o Kappa, que demonstrou maior concord&#226;ncia para a identifica&#231;&#227;o de aus&#234;ncia de les&#227;o (Kappa superior a 90 em todas as compara&#231;&#245;es) e em seguida a identifica&#231;&#227;o de placas calcificadas que tamb&#233;m apresentou um grau de acordo substancial. A concord&#226;ncia diminui na avalia&#231;&#227;o de les&#245;es significativas em ADA (&gt;70%) em que o pior Kappa foi de 47,8 para a avalia&#231;&#227;o interobservador com o software ImageLab.</p>
    <p content-type="asset">
      <table-wrap id="aop08612tab06">
        <graphic xlink:href="1678-4170-abc-99-04-0944-00089-gt06.jpg"/></table-wrap>
    </p>
    <p>
      <bold>Discuss&#227;o</bold>
    </p>
    <p>Nosso trabalho utilizou pacientes encaminhados por seus m&#233;dicos, por indica&#231;&#245;es cl&#237;nicas variadas para a realiza&#231;&#227;o da angiotomografia de coron&#225;rias. Esses pacientes encontrados no dia a dia dos servi&#231;os particulares encaminhados para esse tipo de exame geralmente t&#234;m probabilidade baixa a intermedi&#225;ria de doen&#231;a e utilizam o alto valor preditivo negativo do exame<sup>10</sup> para afastar o diagn&#243;stico. Observamos que na popula&#231;&#227;o estudada, somente 9% apresentavam dor precordial t&#237;pica para DAC, e quando avaliamos o resultado dos escores de c&#225;lcio dessa mesma popula&#231;&#227;o (n&#227;o utilizado nas avalia&#231;&#245;es), somente 20% estavam com valor acima do percentil 75 para sexo e idade, o que est&#225; relacionado com um maior risco global cardiovascular para eventos<sup>11</sup>.</p>
    <p>A utiliza&#231;&#227;o de pacientes encaminhados por indica&#231;&#245;es da pr&#225;tica cardiol&#243;gica permite a viabiliza&#231;&#227;o desse projeto, por&#233;m pode representar um problema. A falta de todos os tipos de les&#227;o e placas pelos segmentos coronarianos certamente &#233; uma limita&#231;&#227;o &#224; an&#225;lise das concord&#226;ncias. Nesses casos s&#243; foi poss&#237;vel realizar a concord&#226;ncia simples, pois o Kappa necessita de uma distribui&#231;&#227;o mais variada dos achados para ser calculado. Uma das formas de corrigir esse problema seria a inclus&#227;o de pacientes distribu&#237;dos pelas les&#245;es e placas, de forma a for&#231;ar a entrada variada de todos os tipos de achados. Essa conduta mudaria o perfil da popula&#231;&#227;o a ser estudada que certamente seria diferente da popula&#231;&#227;o encontrada no dia a dia, encaminhada pelos m&#233;dicos para esclarecimento de uma d&#250;vida diagn&#243;stica. </p>
    <p>A avalia&#231;&#227;o de apenas uma parte do exame (fase contrastada), sem os dados cl&#237;nicos dos pacientes, possivelmente diminui a capacidade dos observadores na interpreta&#231;&#227;o das imagens. Isso pode interferir no laudo do exame e talvez at&#233; na conduta m&#233;dica, por&#233;m nosso objetivo n&#227;o foi avaliar quanto o novo software &#233; acurado em rela&#231;&#227;o ao padr&#227;o de refer&#234;ncia (cineangiocoronariografia), mas sim compar&#225;-lo com outro software amplamente utilizado. </p>
    <p>N&#227;o encontramos em nossa revis&#227;o valida&#231;&#227;o cient&#237;fica espec&#237;fica para o software da Philips, por&#233;m acreditamos que o uso di&#225;rio na pr&#225;tica cl&#237;nica e em trabalhos cient&#237;ficos nos permite concluir que existe algum tipo de valida&#231;&#227;o para esse software.</p>
    <p>Este trabalho foi motivado pela linha de pesquisa financiada pela Faperj de modelagem computacional do sistema cardiovascular. Tem como objetivo futuro utilizar o ImageLab em exames de angiotomografia de coron&#225;rias no formato DICOM e possivelmente outros exames nesse formato.</p>
    <p>A avalia&#231;&#227;o subjetiva do grau de les&#227;o, de uma imagem representativa, de art&#233;ria que est&#225; constantemente em movimento e tem em geral menos que 3 mm pode ser dif&#237;cil. Al&#233;m disso, a visualiza&#231;&#227;o de uma estrutura complexa no espa&#231;o tridimensional expressa de forma bidimensional por uma imagem &#233; um desafio &#224; precis&#227;o. O grau de les&#227;o estimado com essas limita&#231;&#245;es vai definir a conduta cl&#237;nica a ser tomada em v&#225;rios casos.</p>
    <p>Apesar da menor precis&#227;o quando comparada com o cine angiocardiografia (CA), na angiotomografia de coron&#225;rias &#233; poss&#237;vel analisar a imagem em quatro dimens&#245;es (tr&#234;s dimens&#245;es mais o tempo de um ciclo card&#237;aco) e isso possivelmente facilita a concord&#226;ncia com o CA, o que &#233; refletido na alta sensibilidade e especificidade da angiotomografia de coron&#225;rias com valores superiores a 90%<sup>9</sup>. </p>
    <p>Conseguimos uma boa concord&#226;ncia na maioria das an&#225;lises e na avalia&#231;&#227;o direta entre os softwares. Essa concord&#226;ncia foi maior na avalia&#231;&#227;o intraobservador, o que j&#225; era esperado. Outro dado previs&#237;vel foi maior concord&#226;ncia nos casos de aus&#234;ncia de les&#227;o. A pior avalia&#231;&#227;o ocorreu entre os observadores com o software ImageLab na an&#225;lise de les&#227;o &gt;70% na ADA (Kappa de 47,8). Essa discord&#226;ncia poderia ser explicada em parte, pela dif&#237;cil quantifica&#231;&#227;o subjetiva de alguns casos e pelas diferentes interpreta&#231;&#245;es do local em que se encontra a les&#227;o, ou seja, mesmo quando existiu uma avalia&#231;&#227;o correta sobre uma determinada les&#227;o, essa informa&#231;&#227;o n&#227;o era registrada correta se ocorre em segmentos diferentes (les&#227;o &gt; 70% em ADA na intercess&#227;o do segmento proximal e m&#233;dio, ora interpretada com sendo no segmento anterior, ora interpretada como sendo no segmento m&#233;dio).</p>
    <p>Apesar da limita&#231;&#227;o da popula&#231;&#227;o estudada descrita anteriormente, as concord&#226;ncias simples realizadas em casos com poucas observa&#231;&#245;es de les&#227;o e placas foram sempre muito boas. A falta dessa informa&#231;&#227;o pode prejudicar a validade externa do software para achados mais raros, por&#233;m na grande maioria dos pacientes que s&#227;o encaminhados para esse exame n&#227;o se v&#234; essas altera&#231;&#245;es.</p>
    <p>Na an&#225;lise das concord&#226;ncias com maior signific&#226;ncia cl&#237;nica e observa&#231;&#245;es mais frequentes (<xref ref-type="table" rid="aop08612tab06">Tabela 6</xref>) tamb&#233;m observamos maior concord&#226;ncia na observa&#231;&#227;o de aus&#234;ncia de les&#227;o em ADA, com Kappa sempre superior a 90. Nessa mesma tabela se observa menor concord&#226;ncia para les&#227;o &gt;70% e na identifica&#231;&#227;o de placas calcificadas. Esses dados est&#227;o de acordo com as caracter&#237;sticas da angiotomografia de coron&#225;rias, cuja utiliza&#231;&#227;o &#233; mais apropriada para descartar a presen&#231;a les&#245;es e placas, ou seja, o m&#233;todo &#233; mais preciso e &#233; mais confi&#225;vel quando n&#227;o se observa altera&#231;&#245;es coronarianas.</p>
    <p>Quando buscamos na literatura a avalia&#231;&#227;o das concord&#226;ncias com o CAT, apesar da utiliza&#231;&#227;o do padr&#227;o de refer&#234;ncia anat&#244;mico, existe grande varia&#231;&#227;o, com Kappa entre 36 a 63 para medidas dicot&#244;micas detalhadas do grau de les&#227;o (0, 1 - 50,51 - 69 e &gt;70) e Kappa de 37 a 82 para medidas dicot&#244;micas menos detalhadas (&lt;70 ou <underline>&gt;</underline>70)<sup>12</sup>.</p>
    <p>Outro dado interessante utilizando dois observadores &#233; a mudan&#231;a de interpreta&#231;&#227;o sobre um laudo pr&#233;vio de um CAT quando comparado a uma avalia&#231;&#227;o recente, com concord&#226;ncia moderada e Kappa variando entre 54 e 60 entre os observadores. Isso ocorre mesmo quando a concord&#226;ncia recente entre esses observadores &#233; boa (Kappa de 69)<sup>13</sup>. Esses dados podem estar relacionados &#224;s informa&#231;&#245;es dispon&#237;veis pelo observador das imagens durante o momento da an&#225;lise (imagens associadas a hist&#243;ria cl&#237;nica, contexto da realiza&#231;&#227;o do exame, exames pr&#233;vios e sintomas), quando comparada com a an&#225;lise das imagens sem informa&#231;&#227;o pr&#233;via sobre o pacientes.</p>
    <p>No nosso estudo, a aus&#234;ncia de informa&#231;&#245;es sobre o paciente, incluindo a sua identifica&#231;&#227;o, caracter&#237;sticas da dor, precordial, resultado de exames funcionais e a avalia&#231;&#227;o do escore de c&#225;lcio, pode ter influenciado na interpreta&#231;&#227;o das imagens, por&#233;m nosso objetivo n&#227;o foi a libera&#231;&#227;o de um laudo cl&#237;nico, mas sim a compara&#231;&#227;o entre softwares.</p>
    <p>
      <bold>Conclus&#227;o</bold>
    </p>
    <p>Observamos concord&#226;ncias muito boas entre os dois softwares, tanto na avalia&#231;&#227;o intraobservador quanto na avalia&#231;&#227;o interobservador.</p>
    <p>A avalia&#231;&#227;o das concord&#226;ncias intersoftwares, na an&#225;lise de les&#245;es e placas, em geral, foi superior a 80.</p>
    <p>O software ImageLab concorda com o software da Philips na avalia&#231;&#227;o de exames de angiotomografia de coron&#225;rias, tem um grande potencial no aux&#237;lio da pr&#225;tica cl&#237;nica e em pesquisas nessa &#225;rea, apesar da necessidade de uma avalia&#231;&#227;o mais robusta. Pode ser utilizado como ferramenta dentro da linha de pesquisa de modelagem computacional do sistema cardiovascular e possivelmente em outras &#225;reas relacionadas as imagens no formato DICOM. </p>
    <p>
      <bold>Potencial Conflito de Interesses</bold>
    </p>
    <p>Declaro n&#227;o haver conflito de interesses pertinentes.</p>
    <p>
      <bold>Fontes de Financiamento</bold>
    </p>
    <p>O presente estudo n&#227;o teve fontes de financiamento externas.</p>
    <p>
      <bold>Vincula&#231;&#227;o Acad&#234;mica</bold>
    </p>
    <p>Este artigo &#233; parte de disserta&#231;&#227;o de mestrado de Marcelo Souza Hadlich pela Universidade Federal do Rio de Janeiro.</p>
    <p>
      <bold>Refer&#234;ncias</bold>
    </p>
    <p content-type="ref-to-delete">1. Yamauchi T, Yamazaki M, Okawa A,Furuya T, Hayashi K, Sakuma T, et al. Efficacy and reliability of highly functional open source DICOM software (OsiriX) in spine surgery. J Clin Neurosc.. 2010;17(6):756-9. </p>
    <p content-type="ref-to-delete">2. Peanykh OS. A brief history of DICOM. In: Digital Imaging and Communications in Medicine. Philadelphia:Springer;2008.p.17-22. </p>
    <p content-type="ref-to-delete">3. Gibaud B. The DICOM standard: a brief overview. Philadelphia:Springer; 2008.p.1-10.  </p>
    <p content-type="ref-to-delete">4. Graham RNJ, Perriss RW, Scarsbrook AF. DICOM demystified: a review of digital file formats and their use in radiological practice. Clin Radiol. 2005;60(11):1133-40. </p>
    <p content-type="ref-to-delete">5. Kon F. O Software aberto e a quest&#227;o social. Relat&#243;rio T&#233;cnico RT-MAC-2001-07. 2011:1-10.Departamento de Ci&#234;ncia da Computa&#231;&#227;o, IME-USP.[Citado em 2011 jan 10].Dispon&#237;vel em: <ext-link ext-link-type="uri" xlink:href="http://www.ime.usp.br/r Kon/publications.html">http://www.ime.usp.br/r Kon/publications.html</ext-link> </p>
    <p>6. Raff GL, Abidov A, Achenbach S, Berman DS, Boxt LM, Budoff MJ, et al. SCCT guidelines for the interpretation and reporting of&amp;nbsp;coronary computed tomographic angiography. J Cardiovasc Comput Tomogr. 2009;3(2):122-36..</p>
    <p content-type="ref-to-delete">7. Brooks RA. A quantitative theory of the Hounsfield unit and its application to dual energy scanning. J Comput Assist Tomogr. 1977;1(4):487-93. </p>
    <p content-type="ref-to-delete">8. Budoff MJ, Cohen MC, Garcia MJ, Hogson JM, Hundley WG, Lima JA, et al. ACCF/AHA clinical competence statement on cardiac imaging with computed tomography and magnetic resonance: a report of the American College of Cardiology Foundation/American Heart Association/American College of Physicians Task Force on Clinical Competence and Training. J Am Coll Cardiol. 2005;46(2):383-402. </p>
    <p content-type="ref-to-delete">9. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33(1):159-74. </p>
    <p content-type="ref-to-delete">10. Janne d'Oth&#233;e B, Siebert U, Cury R, Jadvar H, Dunn EJ, Hoffmann U. A systematic review on diagnostic accuracy of CT-based detection of significant coronary artery disease. Eur J Radiol. 2008;65(3):449-61. </p>
    <p content-type="ref-to-delete">11. Raggi P, Callister TQ, Cooil B, He ZX, Lippolis NJ, Russo DJ, et al. Identification of patients at increased risk of first unheralded acute myocardial infarction by electron-beam computed tomography. Circulation.2000;101(8):850-5. </p>
    <p content-type="ref-to-delete">12. Guimaraes JA, Victor EG, de Britto Leite MR, Gomes JM, Victor Filho E, Reyes Liveras J. Reliability of the interpretation of coronary angiography by the simple visual method. Arq Bras Cardiol. 2000;74(4):300-8. </p>
    <p content-type="ref-to-delete">13. Banerjee S, Crook AM, Dawson JR, Timmis AD, Hemingway H. Magnitude and Consequences of Error in Coronary Angiography Interpretation (The ACRE Study). Am J Cardiol. 2000;85(3):309-4  </p>
    <p>
      <fn id="end" fn-type="corresp">
        <label>
          <bold> Correspond&#234;ncia:</bold>
        </label>
        <p> Marcelo Souza Hadlich </p>
        <p> Rua Marqu&#234;s de S&#227;o Vicente, 182/102, G&#225;vea </p>
        <p> CEP 22451040, Rio de Janeiro, RJ - Brasil </p>
        <p> E-mail: </p>
        <p><email xlink:href="marcelohadlich@cardiol.br">marcelohadlich@cardiol.br</email>, </p>
        <p>
          <email xlink:href="marcelohadlich@gmail.com">marcelohadlich@gmail.com</email>
        </p>
      </fn>
    </p>
    <p>Artigo recebido em 04/11/11; revisado em 04/11/11; aceito em 13/06/12.</p>
  </body>
  <back>
    <ref-list>
      <ref id="B1">
        <mixed-citation>1. Yamauchi T, Yamazaki M, Okawa A,Furuya T, Hayashi K, Sakuma T, et al. Efficacy and reliability of highly functional open source DICOM software (OsiriX) in spine surgery. J  Clin Neurosc.. 2010;17(6):756-9.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Efficacy and reliability of highly functional open source DICOM software (OsiriX) in spine surgery</article-title>
          <source>J Clin Neurosc</source>
          <date>
            <year>2010</year>
          </date>
          <fpage>756</fpage>
          <lpage>9</lpage>
          <issue>6</issue>
          <volume>17</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Yamauchi</surname>
              <given-names>T</given-names>
            </name>
            <name>
              <surname>Yamazaki</surname>
              <given-names>M</given-names>
            </name>
            <name>
              <surname>Okawa</surname>
              <given-names>A</given-names>
            </name>
            <name>
              <surname>Furuya</surname>
              <given-names>T</given-names>
            </name>
            <name>
              <surname>Hayashi</surname>
              <given-names>K</given-names>
            </name>
            <name>
              <surname>Sakuma</surname>
              <given-names>T</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B2">
        <mixed-citation>2. Peanykh OS. A brief history of DICOM. In: Digital Imaging and Communications in Medicine. Philadelphia:Springer;2008.p.17-22.</mixed-citation>
        <element-citation publication-type="book">
          <source>Digital Imaging and Communications in Medicine</source>
          <date>
            <year>2008</year>
          </date>
          <fpage>17</fpage>
          <lpage>22</lpage>
          <person-group person-group-type="author">
            <name>
              <surname>Peanykh</surname>
              <given-names>OS</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B3">
        <mixed-citation>3. Gibaud B. The DICOM standard: a brief overview. Philadelphia:Springer; 2008.p.1-10.</mixed-citation>
        <element-citation publication-type="book">
          <source>The DICOM standard: a brief overview</source>
          <date>
            <year>2008</year>
          </date>
          <person-group person-group-type="author">
            <name>
              <surname>Gibaud</surname>
              <given-names>B</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B4">
        <mixed-citation>4. Graham RNJ, Perriss RW, Scarsbrook AF. DICOM demystified: a review of digital file formats and their use in radiological practice. Clin Radiol. 2005;60(11):1133-40.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>DICOM demystified: a review of digital file formats and their use in radiological practice</article-title>
          <source>Clin Radiol</source>
          <date>
            <year>2005</year>
          </date>
          <fpage>1133</fpage>
          <lpage>40</lpage>
          <issue>11</issue>
          <volume>60</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Graham</surname>
              <given-names>RNJ</given-names>
            </name>
            <name>
              <surname>Perriss</surname>
              <given-names>RW</given-names>
            </name>
            <name>
              <surname>Scarsbrook</surname>
              <given-names>AF</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B5">
        <mixed-citation>5. Kon F. O Software aberto e a quest&#227;o social. Relat&#243;rio T&#233;cnico RT-MAC-2001-07. 2011:1-10.Departamento de Ci&#234;ncia da Computa&#231;&#227;o, IME-USP.[Citado em 2011 jan 10].Dispon&#237;vel em: <ext-link ext-link-type="uri" xlink:href="http://www.ime.usp.br/r  Kon/publications.html">http://www.ime.usp.br/r  Kon/publications.html</ext-link></mixed-citation>
        <element-citation publication-type="book">
          <source>O Software aberto e a quest&#227;o social: Relat&#243;rio T&#233;cnico RT-MAC-2001-07</source>
          <date>
            <year>2011</year>
          </date>
          <fpage>1</fpage>
          <lpage>10</lpage>
          <person-group person-group-type="author">
            <name>
              <surname>Kon</surname>
              <given-names>F</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B6">
        <element-citation publication-type="journal">
          <article-title>SCCT guidelines for the interpretation and reporting of coronary computed tomographic angiography</article-title>
          <source>J Cardiovasc Comput Tomogr</source>
          <date>
            <year>2009</year>
          </date>
          <fpage>122</fpage>
          <lpage>36</lpage>
          <issue>2</issue>
          <volume>3</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Raff</surname>
              <given-names>GL</given-names>
            </name>
            <name>
              <surname>Abidov</surname>
              <given-names>A</given-names>
            </name>
            <name>
              <surname>Achenbach</surname>
              <given-names>S</given-names>
            </name>
            <name>
              <surname>Berman</surname>
              <given-names>DS</given-names>
            </name>
            <name>
              <surname>Boxt</surname>
              <given-names>LM</given-names>
            </name>
            <name>
              <surname>Budoff</surname>
              <given-names>MJ</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B7">
        <mixed-citation>7. Brooks RA. A quantitative theory of the Hounsfield unit and its application to dual energy scanning. J Comput Assist Tomogr. 1977;1(4):487-93.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>A quantitative theory of the Hounsfield unit and its application to dual energy scanning</article-title>
          <source>J Comput Assist Tomogr</source>
          <date>
            <year>1977</year>
          </date>
          <fpage>487</fpage>
          <lpage>93</lpage>
          <issue>4</issue>
          <volume>1</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Brooks</surname>
              <given-names>RA</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B8">
        <mixed-citation>8. Budoff MJ, Cohen MC, Garcia MJ, Hogson JM, Hundley WG, Lima JA, et al. ACCF/AHA clinical competence statement on cardiac imaging with computed tomography and magnetic resonance: a report of the American College of Cardiology Foundation/American Heart Association/American College of Physicians Task Force on Clinical Competence and Training. J Am Coll Cardiol. 2005;46(2):383-402.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>ACCF/AHA clinical competence statement on cardiac imaging with computed tomography and magnetic resonance: a report of the American College of Cardiology Foundation/American Heart Association/American College of Physicians Task Force on Clinical Competence and Training</article-title>
          <source>J Am Coll Cardiol</source>
          <date>
            <year>2005</year>
          </date>
          <fpage>383</fpage>
          <lpage>402</lpage>
          <issue>2</issue>
          <volume>46</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Budoff</surname>
              <given-names>MJ</given-names>
            </name>
            <name>
              <surname>Cohen</surname>
              <given-names>MC</given-names>
            </name>
            <name>
              <surname>Garcia</surname>
              <given-names>MJ</given-names>
            </name>
            <name>
              <surname>Hogson</surname>
              <given-names>JM</given-names>
            </name>
            <name>
              <surname>Hundley</surname>
              <given-names>WG</given-names>
            </name>
            <name>
              <surname>Lima</surname>
              <given-names>JA</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B9">
        <mixed-citation>9. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33(1):159-74.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>The measurement of observer agreement for categorical data</article-title>
          <source>Biometrics</source>
          <date>
            <year>1977</year>
          </date>
          <fpage>159</fpage>
          <lpage>74</lpage>
          <issue>1</issue>
          <volume>33</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Landis</surname>
              <given-names>JR</given-names>
            </name>
            <name>
              <surname>Koch</surname>
              <given-names>GG</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B10">
        <mixed-citation>10. Janne d'Oth&#233;e B, Siebert U, Cury R, Jadvar H, Dunn EJ, Hoffmann U. A systematic review on diagnostic accuracy of CT-based detection of significant coronary artery disease. Eur J Radiol. 2008;65(3):449-61.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>A systematic review on diagnostic accuracy of CT-based detection of significant coronary artery disease</article-title>
          <source>Eur J Radiol</source>
          <date>
            <year>2008</year>
          </date>
          <fpage>449</fpage>
          <lpage>61</lpage>
          <issue>3</issue>
          <volume>65</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Janne d'Oth&#233;e</surname>
              <given-names>B</given-names>
            </name>
            <name>
              <surname>Siebert</surname>
              <given-names>U</given-names>
            </name>
            <name>
              <surname>Cury</surname>
              <given-names>R</given-names>
            </name>
            <name>
              <surname>Jadvar</surname>
              <given-names>H</given-names>
            </name>
            <name>
              <surname>Dunn</surname>
              <given-names>EJ</given-names>
            </name>
            <name>
              <surname>Hoffmann</surname>
              <given-names>U</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B11">
        <mixed-citation>11. Raggi P, Callister TQ, Cooil B, He ZX, Lippolis NJ, Russo DJ, et al. Identification of patients at increased risk of first unheralded acute myocardial infarction by electron-beam computed tomography. Circulation.2000;101(8):850-5.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Identification of patients at increased risk of first unheralded acute myocardial infarction by electron-beam computed tomography</article-title>
          <source>Circulation</source>
          <date>
            <year>2000</year>
          </date>
          <fpage>850</fpage>
          <lpage>5</lpage>
          <issue>8</issue>
          <volume>101</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Raggi</surname>
              <given-names>P</given-names>
            </name>
            <name>
              <surname>Callister</surname>
              <given-names>TQ</given-names>
            </name>
            <name>
              <surname>Cooil</surname>
              <given-names>B</given-names>
            </name>
            <name>
              <surname>He</surname>
              <given-names>ZX</given-names>
            </name>
            <name>
              <surname>Lippolis</surname>
              <given-names>NJ</given-names>
            </name>
            <name>
              <surname>Russo</surname>
              <given-names>DJ</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B12">
        <mixed-citation>12. Guimaraes JA, Victor EG, de Britto Leite MR, Gomes JM, Victor Filho E, Reyes Liveras J. Reliability of the interpretation of coronary angiography by the simple visual method. Arq Bras Cardiol. 2000;74(4):300-8.</mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Reliability of the interpretation of coronary angiography by the simple visual method</article-title>
          <source>Arq Bras Cardiol</source>
          <date>
            <year>2000</year>
          </date>
          <fpage>300</fpage>
          <lpage>8</lpage>
          <issue>4</issue>
          <volume>74</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Guimaraes</surname>
              <given-names>JA</given-names>
            </name>
            <name>
              <surname>Victor</surname>
              <given-names>EG</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
      <ref id="B13">
        <mixed-citation>13.  Banerjee S,  Crook AM, Dawson JR,  Timmis AD, Hemingway H.  Magnitude and Consequences of Error in Coronary Angiography Interpretation (The ACRE Study). Am J Cardiol. 2000;85(3):309-4 </mixed-citation>
        <element-citation publication-type="journal">
          <article-title>Magnitude and Consequences of Error in Coronary Angiography Interpretation (The ACRE Study)</article-title>
          <source>Am J Cardiol</source>
          <date>
            <year>2000</year>
          </date>
          <fpage>309</fpage>
          <lpage>4</lpage>
          <issue>3</issue>
          <volume>85</volume>
          <person-group person-group-type="author">
            <name>
              <surname>Banerjee</surname>
              <given-names>S</given-names>
            </name>
            <name>
              <surname>Crook</surname>
              <given-names>AM</given-names>
            </name>
            <name>
              <surname>Dawson</surname>
              <given-names>JR</given-names>
            </name>
            <name>
              <surname>Timmis</surname>
              <given-names>AD</given-names>
            </name>
            <name>
              <surname>Hemingway</surname>
              <given-names>H</given-names>
            </name>
          </person-group>
        </element-citation>
      </ref>
    </ref-list>
  </back>
  <sub-article article-type="translation" id="TRen" xml:lang="en">
    <front-stub>
      <article-categories>
        <subj-group subj-group-type="heading">
          <subject>Original Articles</subject>
        </subj-group>
      </article-categories>
      <title-group>
        <article-title xml:lang="en">Free and open-source software application for the evaluation of coronary computed tomography angiography images</article-title>
      </title-group>
      <abstract xml:lang="en">
        <p>BACKGROUND: The standardization of images used in Medicine in 1993 was performed using the DICOM (Digital Imaging and Communications in Medicine) standard. Several tests use this standard and it is increasingly necessary to design software applications capable of handling this type of image; however, these software applications are not usually free and open-source, and this fact hinders their adjustment to most diverse interests. OBJECTIVE: To develop and validate a free and open-source software application capable of handling DICOM coronary computed tomography angiography images. METHODS: We developed and tested the ImageLab software in the evaluation of 100 tests randomly selected from a database. We carried out 600 tests divided between two observers using ImageLab and another software sold with Philips Brilliance computed tomography appliances in the evaluation of coronary lesions and plaques around the left main coronary artery (LMCA) and the anterior descending artery (ADA). To evaluate intraobserver, interobserver and intersoftware agreements, we used simple and kappa statistics agreements. RESULTS: The agreements observed between software applications were generally classified as substantial or almost perfect in most comparisons. CONCLUSION: The ImageLab software agreed with the Philips software in the evaluation of coronary computed tomography angiography tests, especially in patients without lesions, with lesions &lt; 50% in the LMCA and &lt; 70% in the ADA. The agreement for lesions &gt; 70% in the ADA was lower, but this is also observed when the anatomical reference standard is used.</p>
      </abstract>
      <kwd-group xml:lang="en">
        <kwd>Diagnostic imaging</kwd>
        <kwd>tomography scanners</kwd>
        <kwd>X-Ray computed</kwd>
        <kwd>coronary vessels</kwd>
        <kwd>software validation</kwd>
      </kwd-group>
    </front-stub>
    <body>
      <p>
        <bold>Free and open-source software application for the evaluation of coronary computed tomography angiography images</bold>
      </p>
      <p>
        <bold>Marcelo Souza Hadlich<sup>I,II,IV,V,VI</sup>; Gl&#225;ucia Maria Moraes Oliveira<sup>I,II,IV</sup>; Ra&#250;l A. Feij&#243;o<sup>III</sup>; Clerio F. Azevedo<sup>IV,V,VI</sup>; Bernardo Rangel Tura<sup>V</sup>; Paulo Gustavo Portela Ziemer<sup>III</sup>; Pablo Javier Blanco<sup>III</sup>; Gustavo Pina<sup>I</sup>; M&#225;rcio Meira<sup>I</sup>; Nelson Albuquerque de Souza e Silva<sup>I,II</sup></bold>
      </p>
      <p><sup>I</sup>Universidade Federal do Rio de Janeiro (UFRJ). Rio de Janeiro, RJ - Brazil</p>
      <p><sup>II</sup>Instituto do Cora&#231;&#227;o Edson Saad. Rio de Janeiro, RJ - Brazil</p>
      <p><sup>III</sup>Laborat&#243;rio Nacional de Computa&#231;&#227;o Cient&#237;fica (LNCC). Rio de Janeiro, RJ - Brazil</p>
      <p><sup>IV</sup>Rede Labs Dor. Rio de Janeiro, RJ - Brazil</p>
      <p><sup>V</sup>Instituto Nacional de Cardiologia. Rio de Janeiro, RJ - Brazil</p>
      <p><sup>VI</sup>Instituto Dor de Pesquisa e Ensino. Rio de Janeiro, RJ - Brazil</p>
      <p>
        <xref ref-type="corresp" rid="end-body2">Mailing Address</xref>
      </p>
      <p>
        <bold>ABSTRACT</bold>
      </p>
      <p><bold>BACKGROUND: </bold>The standardization of images used in Medicine in 1993 was performed using the DICOM (Digital Imaging and Communications in Medicine) standard. Several tests use this standard and it is increasingly necessary to design software applications capable of handling this type of image; however, these software applications are not usually free and open-source, and this fact hinders their adjustment to most diverse interests.</p>
      <p><bold>OBJECTIVE:</bold> To develop and validate a free and open-source software application capable of handling DICOM coronary computed tomography angiography images.</p>
      <p><bold>METHODS:</bold> We developed and tested the ImageLab software in the evaluation of 100 tests randomly selected from a database. We carried out 600 tests divided between two observers using ImageLab and another software sold with Philips Brilliance computed tomography appliances in the evaluation of coronary lesions and plaques around the left main coronary artery (LMCA) and the anterior descending artery (ADA). To evaluate intraobserver, interobserver and intersoftware agreements, we used simple and kappa statistics agreements.</p>
      <p><bold>RESULTS: </bold> The agreements observed between software applications were generally classified as substantial or almost perfect in most comparisons.</p>
      <p><bold>CONCLUSION: </bold> The ImageLab software agreed with the Philips software in the evaluation of coronary computed tomography angiography tests, especially in patients without lesions, with lesions &lt; 50% in the LMCA and &lt; 70% in the ADA. The agreement for lesions &gt; 70% in the ADA was lower, but this is also observed when the anatomical reference standard is used.</p>
      <p><bold> Keywords:</bold> Diagnostic imaging; tomography scanners, X-Ray computed; coronary vessels; software validation.</p>
      <p>
        <bold>Introduction</bold>
      </p>
      <p>In the medical area, a specific imaging pattern called DICOM (Digital Imaging and Communications in Medicine) is used. This standard was developed to facilitate communication between the software and hardware associated with this process.</p>
      <p>The DICOM format, standardized<sup>1</sup> in 1993 by the RSNA (Radiological Society of North America) Congress, aimed at standardizing the rules with which medical information is transmitted and stored<sup>2-4</sup>.</p>
      <p>Medical images are viewed and processed by means of specific software applications. Many of these resources are not available for free or are sold together with equipment, but it is not common to find free and open-source software applications for that image format. The public domain of a medical software technology could reduce medical costs, homogenize multicenter use, allow continued development and facilitate the development of several lines of research, and some believe that this format could contribute to reducing the social gap between countries<sup>5</sup>.</p>
      <p>The National Institute of Science and Technology in Medicine Assisted by Scientific Computing (INCT-MACC) involves 33 national institutions from 11 states, totaling 128 researchers. One area of research of the INCT-MACC is the processing of medical images, enabling the acquisition of information to improve the computational modeling of the human cardiovascular system, also under development by the INCT-MACC. For this purpose, a free and open-source software application named ImageLab was designed. The characteristics of this software supposedly include the fact that it is user-friendly, but that requires further evaluation.</p>
      <p>The purpose of the study was to develop and evaluate reliability, reproducibility and agreement of the new software when compared to another being currently used in coronary computed tomography angiography equipment.</p>
      <p>
        <bold>Methods</bold>
      </p>
      <p>In a concerted effort undertaken by the Universidade Federal do Rio de Janeiro (UFRJ) together with the National Laboratory for Scientific Computing (LNCC), a software application was developed to analyze DICOM images. This free and open-source software is called ImageLab and has some tools created to analyze and process coronary computed tomography angiography images.</p>
      <p>The inclusion criteria was applied to a population of 6,216 patients who underwent coronary computed tomography angiography tests between May 2005 and December 2010, i.e., patients with complete data recorded in the database created in 2008, of which 2,895 patients were selected. Subsequently, after the exclusion criteria were applied, i.e., presence of stents in the left main coronary artery or anterior descending artery (ADA), surgical clip, resulting from coronary artery bypass grafting and patients who had no images stored in the server, a total of 534 cases were selected (<xref ref-type="fig" rid="en_aop08612fig01-body2">Figure 1</xref>).</p>
      <p content-type="asset">
        <fig id="en_aop08612fig01-body2">
          <graphic xlink:href="1678-4170-abc-99-04-0944-00089-gf01-en.jpg"/></fig>
      </p>
      <p>A random selection was carried out in the randon.org website, which removed 120 of the 534 patients, and the first 100 cases were then used. All patients had their identifiers replaced by consecutive numbers, by people not involved in the image analysis study, making it blind.</p>
      <p>Raw data from coronary computed tomography angiography images obtained through CT scanners available in the service were used (GE Lightspeed VTC 64 and 32, Philips Brilliance 64, 40 and 16). All images followed a standard acquisition protocol using similar techniques.</p>
      <p>All 100 tests were analyzed by two software applications. The Philips Brilliance software was purchased along with the cardiovascular package (available for a specific workstation for analyses of DICOM images), and the ImageLab software, developed in this project, was installed in the same workstation (to neutralize potential variations of processing capacity if they were installed in computers with different capacities).</p>
      <p>We chose the left main coronary artery and the ADA due to their clinical importance and to enable the analysis, as the use of the 16 segments in their entirety would result in a large number of variables. We followed the nomenclature proposed by the AHA, regarding segments 5, 6, 7 and 8<sup>6</sup>. The evaluation of the presence of coronary lesions greater than or smaller than 50% was used for segment 5 (left main coronary artery) and for the other segments (anterior descending artery), we used cutoff points of 70%. Additionally, the following variables were also recorded: Presence of calcified, partially calcified (in each segment) and non-calcified atherosclerotic plaques; subjective quality of images; time of image analysis (recorded by observer 1).</p>
      <p>Image analysis was performed by viewing orthogonal planes selected from image blocks in three-dimensional space, starting with the chest analysis in the transverse plane from head to foot, in order to view the aorta, the origin of the left main coronary artery and the ADA. Evaluations on the degree of stenosis followed the standard analysis, where the filling failure in the contrasted vessel defines the lesion. The analysis of the type of plaque responsible for the lesions was performed subjectively. Plaque characterization was performed according to their morphology and signal intensity following the traditional scale adopted in CT tests, the scale of Hounsfield (HU)<sup>7</sup>. The plaques were classified as: calcified (tissue adjacent to the vessel with more signal strength than the contrasted vessel - signal &gt;130 HU); non-calcified (tissue adjacent to the vessel with lower signal than the contrasted vessel - signal &lt; -50 HU), and partially calcified (heterogeneous content).</p>
      <p>The two software applications feature a tool designed to find and analyze any points in the three-dimensional space. This tool was used to evaluate the vessel in its axial plane (viewing from within the vessels) throughout their extension, and when coronary lesions and plaques are present, all the orthogonal planes may be used simultaneously at the exact topography where these findings are located.</p>
      <p>
        <bold>Analysis of the Images</bold>
      </p>
      <p>Two observers with more than five years of experience, equivalent to level 3 of clinical competence<sup>8</sup>, used the commercial Philips software and ImageLab, installed on the same computer, to analyze the segment of the ADA according to the variables described in <xref ref-type="table" rid="tab1-body2">Table 1</xref>. To avoid bias, the analyses of the two software applications occurred at intervals of more than 15 days.</p>
      <table-wrap id="tab1-body2">
        <graphic xlink:href="1678-4170-abc-99-04-0944-00089-gt01-en.jpg"/></table-wrap>
      <p>A total of 600 analyses were performed to assess the intraobserver, interobserver and intersoftware agreements.</p>
      <p>
        <bold>Statistical Analysis</bold>
      </p>
      <p>Descriptive analysis of numerical variables used means +/- standard deviations. Categorical variables were analyzed as numbers (n) and percentages (%). Kappa statistics were used to calculate the interobserver and intraobserver reproducibility, all with a confidence interval of 95%, as well as simple agreement to measure the cases with a small number of disagreements. The software used was R for Linux. We used <xref ref-type="table" rid="tab1-body2">Table 1</xref> to assess the degree of consistency between the agreements<sup>9</sup>. </p>
      <p>
        <bold>Results</bold>
      </p>
      <p>After the random selection of 120 cases, 7 cases that scored less than 100 were excluded due to lack of contrast examination (calcium score tests without coronary computed tomography angiography). A total of 100 tests that scored 1-107 were included in the 600 analyses divided by observers and software. (All results of agreements obtained with the R program for Linux can be viewed at <ext-link ext-link-type="uri" xlink:href="http://cl.ly/3c193E0J1o1M0f360u1d">http://cl.ly/3c193E0J1o1M0f360u1d</ext-link>)</p>
      <p>The results of the ImageLab software version used in this project showed to be appropriate, demonstrating convenient format and usability (<xref ref-type="fig" rid="en_aop08612fig02-body2">Figure 2</xref>).</p>
      <p content-type="asset">
        <fig id="en_aop08612fig02-body2">
          <graphic xlink:href="1678-4170-abc-99-04-0944-00089-gf02-en.jpg"/></fig>
      </p>
      <p>All 100 patients analyzed were tested between July 2009 and November 2010, most of whom were male (65) and mean age was 58 years (23-85). Mean body mass index was 27 kg/m<sup>2</sup>, which classified the population as overweight (25% were classified as obese). Only 15% of the population did not have any risk factor, and 35% were asymptomatic (<xref ref-type="table" rid="tab2-body2">Table 2</xref>).</p>
      <table-wrap id="tab2-body2">
        <graphic xlink:href="1678-4170-abc-99-04-0944-00089-gt02-en.jpg"/></table-wrap>
      <p>The most frequent symptom observed was the presence of atypical pain for CAD, and most test indications were performed to evaluate any symptoms. It is noteworthy that 36% of patients were referred due to the presence of abnormal functional test, and 16%, for risk stratification (routine evaluation unrelated to symptoms or abnormal tests); only 2%, to exclude the diagnosis of atherothrombotic disease as the etiology of a cardiomyopathy; and 2%, due to preoperative evaluation (assessment of surgical cardiovascular risk).</p>
      <p>Observer 1 used a pad-shaped electronic device (Apple<sup>&#174;</sup> tablet - Wi-fi 64 Gb iPad) to record the analyses using the software application Bento<sup>&#174;</sup> for iPad. This allowed registering the time of the analyses in a simple way (<xref ref-type="table" rid="en_aop08612tab03-body2">Table 3</xref>).</p>
      <p content-type="asset">
        <table-wrap id="en_aop08612tab03-body2">
          <graphic xlink:href="1678-4170-abc-99-04-0944-00089-gt03-en.jpg"/></table-wrap>
      </p>
      <p>The total time of the analyses performed by observer 1 was approximately 6.8 hours, throughout 14 days. The average time spent with each analysis was 2 minutes and 4 seconds (21 to 612 seconds). The mean time using ImageLab software to analyze all cases was 226.2 minutes, while with the software Philips Brilliance, it was 180.9 (20% less). The second analysis using both software applications was faster with the decrease in total time by 11.9 minutes and 10.7 minutes for the software applications ImageLab and Philips Brilliance, respectively.</p>
      <p>We recorded the subjective quality of images through a 1-3 score, 1 - low quality, 2 - intermediate quality, and 3 - high quality. Twenty-four percent of the images were classified as low quality; 64.2% as intermediate, and 11.8%, as high quality (76% with intermediate or high quality).</p>
      <p>To evaluate the intraobserver and interobserver agreements, we used the simple agreement (sum of agreeing findings in relation to the total number of cases), in the presence of few or no cases of disagreement (heterogeneous distribution in the 2 x 2 table with few or no cases in one of the cells), and using kappa statistics (<xref ref-type="table" rid="tab1-body2">Table 1</xref>) as a reference to the degree of agreement between the results<sup>8</sup>. </p>
      <p>The average observations of the 600 analyzes (<xref ref-type="table" rid="en_aop08612tab03-body2">Table 3</xref>) showed no lesion in the main coronary artery in 82%, and in the ADA in 49.3%. Less than 1% had lesions &gt;50% in the main coronary artery and 9% had lesions &gt;70% in some segment of the ADA. The most frequently observed plaque type was the calcified one. </p>
      <p>It was possible to perform the analysis of agreement using Kappa, for all types of comparison (intraobserver, interobserver, and intersoftware), in the following variables: absence of lesion and lesion &gt; 50% in the left main coronary artery; absence of lesion in the ADA and &gt; 70% in the middle third of this artery (<xref ref-type="table" rid="en_aop08612tab04-body2">Table 4</xref>). For the evaluation of plaques, Kappa may be performed only with the variable calcified plaque in the left main coronary artery (<xref ref-type="table" rid="en_aop08612tab05-body2">Table 5</xref>). The other agreements were performed by measuring the simple agreement.</p>
      <p content-type="asset">
        <table-wrap id="en_aop08612tab04-body2">
          <graphic xlink:href="1678-4170-abc-99-04-0944-00089-gt04-en.jpg"/></table-wrap>
      </p>
      <p content-type="asset">
        <table-wrap id="en_aop08612tab05-body2">
          <graphic xlink:href="1678-4170-abc-99-04-0944-00089-gt05-en.jpg"/></table-wrap>
      </p>
      <p>In order to evaluate the intraobserver agreements, we compared the analyses performed by observer 1 with the software ImageLab at time 1 ImageLab with time 2 as well as the Philips Brilliance software at time 1 Philips Brilliance at time 2.</p>
      <p>All evaluations measured by Kappa or by simple agreement in left main coronary artery were &gt;60. Only the variable lesion &gt;70% in the mid third of the descending artery, in the interobserver evaluation by Philips Brilliance, presented moderate agreement (49.7 - 13.8 to 85.6). All the other agreements observed in the left ADA were &gt;60<sup>9</sup>. </p>
      <p>The observation of absence of lesions and lesion &lt;50% in the left main coronary artery, as well as lesion &gt;70% in the mid-third of the left ADA showed greater intraobserver agreement by the software Philips Brilliance. The others were similar.</p>
      <p>The intersoftware analysis showed agreement &gt;80 for all measurements performed by Kappa, except for the observation of lesion &gt;70% in the mid-third of the ADA (79.6 - 63.6 to 95.7), while all simple intersoftware agreements were &gt;80.</p>
      <p>When the most frequent observations with greater clinical significance are analyzed (<xref ref-type="table" rid="en_aop08612tab06-body2">Table 6</xref>), it was possible to use the Kappa, which showed higher agreement for the identification of absence of lesion (Kappa &gt;90 for all comparisons) and then, the identification of calcified plaques that also displayed a substantial degree of agreement. The agreement decreases in the evaluation of significant lesions in the ADA (&gt;70%), in which the worst kappa was 47.8 for the interobserver evaluation with the ImageLab software.</p>
      <p content-type="asset">
        <table-wrap id="en_aop08612tab06-body2">
          <graphic xlink:href="1678-4170-abc-99-04-0944-00089-gt06-en.jpg"/></table-wrap>
      </p>
      <p>
        <bold>Discussion</bold>
      </p>
      <p>Our study analyzed patients referred by their physicians to undergo coronary computed tomography angiography due to different clinical indications. These patients, seen at the routine of private services referred to these types of tests, generally have low to intermediate likelihood of having the disease and the high negative predictive value of the test<sup>10</sup> is used to rule out the diagnosis. We observed that in the population studied, only 9% had chest pain typical of coronary artery disease and when we evaluated the results of calcium scores of the same population (not used in the assessments), only 20% had values above the 75<sup>th</sup> percentile for sex and age, which is associated with a higher overall risk for cardiovascular events<sup>11</sup>.</p>
      <p>The use of patients referred for cardiological indications allows the feasibility of this study, but it may also pose a problem. The lack of all types of lesions and plaques by coronary segments surely is a limitation for the analysis of agreements. In these cases, it was only possible to perform the simple agreement, as the Kappa needs a more varied distribution of findings to be calculated. One way to correct this problem would be to include patients distributed by lesions and plaques in order to force the varied entry of all kinds of findings. This approach would change the profile of the population to be studied, which would certainly be different from the population seen in everyday life, which is referred by physicians for clarification of an uncertain diagnosis. </p>
      <p>The assessment of only part of the test (contrasted phase) without the patients' clinical data, possibly diminishes the observers' capacity to interpret the images. This may interfere with the test report and perhaps with medical management, but our goal was not to evaluate how accurate the new software was in relation to the reference standard (coronary angiography), but to compare it with another widely used software application. </p>
      <p>We did not find any specific scientific validation for the Philips' software in our review, but we believe that its regular use in clinical practice and in scientific works allows us to conclude that there is some sort of validation for that software application.</p>
      <p>This study was motivated by the research area funded by Faperj, concerning computer modeling of the cardiovascular system. It aims to use ImageLab in coronary angiography tests in DICOM format and possibly in other tests using this format.</p>
      <p>The subjective evaluation of the degree of lesion, from a representative image of the artery that is constantly in motion and is generally smaller than 3 mm, can be difficult. Moreover, viewing a complex structure in three-dimensional space expressed by a two-dimensional shape image is a challenge to accuracy. The lesion degree estimated with these limitations will define the clinical conduct to be followed in several cases.</p>
      <p>Despite the lower accuracy compared with cine angiocardiography (CA), it is possible to analyze the image in four dimensions (three dimensions plus the time of one cardiac cycle) when using coronary computed tomography angiography and that possibly facilitates agreement with CA, which is reflected in the high sensitivity and specificity of coronary computed tomography angiography, with values above 90%<sup>9</sup>.</p>
      <p>We obtained good agreement in most analyses and in the direct evaluation of the software applications. This agreement was greater in the intraobserver evaluation, which was expected. Another predictable finding was greater agreement in cases of absence of lesion. The worst evaluation occurred between observers with the software ImageLab in the analysis of lesion &gt;70% in the ADA (Kappa 47.8). This discrepancy could be partly explained by the difficult subjective quantification in some cases and the different interpretations of the site where the lesion is found, or even if there was a correct assessment of a particular lesion, as this information was not accurately recorded to state whether it occurred in different segments (lesion &gt;70% in the ADA at the intersection of the proximal and middle segments, sometimes interpreted as in the anterior segment, sometimes interpreted as in the middle segment).</p>
      <p>Despite the limitations of the study population described above, the simple agreements performed in cases with few observations of lesion and plaques were always very good. The lack of such information may jeopardize the external validity of the software for rarer findings, but the vast majority of patients who are referred to this test do not have these alterations.</p>
      <p>In the analysis of agreements with greater clinical significance and more frequent observations (<xref ref-type="table" rid="en_aop08612tab06-body2">Table 6</xref>) we also observed greater agreement for the observation of absence of lesions in the ADA, with Kappa always &gt;90. In this same table, we observed lower agreement for lesion &gt;70% and in the identification of calcified plaques. These data are consistent with the characteristics of coronary angiography, of which use is more appropriate to rule out the presence of lesions and plaques, i.e., the method is more accurate and is more reliable when coronary alterations are observed.</p>
      <p>When we performed a search in the literature for evaluation of agreements with CA, although the anatomical reference standard is used, we found a great variation, with Kappa ranging from 36 to 63 for detailed dichotomous measures on the degree of lesion (0.1 - 50.51 - 69 and &gt;70) and Kappa of 37 to 82 for less detailed dichotomous measures (&gt;70 or &gt;70)<sup>12</sup>.</p>
      <p>Another interesting data using two observers is the change of interpretation in a previous report of CAT compared to a recent evaluation, with moderate agreement and Kappa ranging from 54 to 60 among observers. This occurs even when the recent agreement between these observers is good (Kappa 69)<sup>13</sup>. These data may be related to the information available for the image observer during the time of analysis (images associated with clinical history, the test context, previous examinations and symptoms), compared to the analysis of images without prior information about the patients.</p>
      <p>In our study, the absence of patient information, including identification, characteristics of precordial pain, result of functional tests and evaluation of calcium score, may have influenced the interpretation of images, but our goal was not to approve a clinical report, but to compare software applications.</p>
      <p>
        <bold>Conclusion</bold>
      </p>
      <p>We observed very good agreements between the two software applications, both in the intraobserver and the interobserver evaluation.</p>
      <p>The evaluation of intersoftware agreements, analysis of lesions and plaques was generally greater than 80.</p>
      <p>The software ImageLab agrees with the Philips software in the evaluation of coronary computed tomography angiography, has a great potential to help clinical practice and research in this area, despite the need for a more robust evaluation. It can be used as a tool in the research area of computer modeling of the cardiovascular system and possibly in other areas related to images in DICOM format.</p>
      <p>
        <bold>Potential Conflict of Interest</bold>
      </p>
      <p>No potential conflict of interest relevant to this article was reported.</p>
      <p>
        <bold>Sources of Funding</bold>
      </p>
      <p>There were no external funding sources for this study.</p>
      <p>
        <bold>Study Association</bold>
      </p>
      <p>This article is part of the dissertation of master submitted by Marcelo Souza Hadlich, from Universidade Federal do Rio de Janeiro.</p>
      <p>
        <bold>References</bold>
      </p>
      <p content-type="ref-to-delete">1. Yamauchi T, Yamazaki M, Okawa A,Furuya T, Hayashi K, Sakuma T, et al. Efficacy and reliability of highly functional open source DICOM software (OsiriX) in spine surgery. J Clin Neurosc.. 2010;17(6):756-9. </p>
      <p content-type="ref-to-delete">2. Peanykh OS. A brief history of DICOM. In: Digital Imaging and Communications in Medicine. Philadelphia:Springer;2008.p.17-22. </p>
      <p content-type="ref-to-delete">3. Gibaud B. The DICOM standard: a brief overview. Philadelphia:Springer; 2008.p.1-10.  </p>
      <p content-type="ref-to-delete">4. Graham RNJ, Perriss RW, Scarsbrook AF. DICOM demystified: a review of digital file formats and their use in radiological practice. Clin Radiol. 2005;60(11):1133-40. </p>
      <p content-type="ref-to-delete">5. Kon F. O Software aberto e a quest&#227;o social. Relat&#243;rio T&#233;cnico RT-MAC-2001-07. 2011:1-10.Departamento de Ci&#234;ncia da Computa&#231;&#227;o, IME-USP.[Citado em 2011 jan 10].Dispon&#237;vel em: <ext-link ext-link-type="uri" xlink:href="http://www.ime.usp.br/r Kon/publications.html">http://www.ime.usp.br/r Kon/publications.html</ext-link> </p>
      <p>6. Raff GL, Abidov A, Achenbach S, Berman DS, Boxt LM, Budoff MJ, et al. SCCT guidelines for the interpretation and reporting of&amp;nbsp;coronary computed tomographic angiography. J Cardiovasc Comput Tomogr. 2009;3(2):122-36..</p>
      <p content-type="ref-to-delete">7. Brooks RA. A quantitative theory of the Hounsfield unit and its application to dual energy scanning. J Comput Assist Tomogr. 1977;1(4):487-93. </p>
      <p content-type="ref-to-delete">8. Budoff MJ, Cohen MC, Garcia MJ, Hogson JM, Hundley WG, Lima JA, et al. ACCF/AHA clinical competence statement on cardiac imaging with computed tomography and magnetic resonance: a report of the American College of Cardiology Foundation/American Heart Association/American College of Physicians Task Force on Clinical Competence and Training. J Am Coll Cardiol. 2005;46(2):383-402. </p>
      <p content-type="ref-to-delete">9. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33(1):159-74. </p>
      <p content-type="ref-to-delete">10. Janne d'Oth&#233;e B, Siebert U, Cury R, Jadvar H, Dunn EJ, Hoffmann U. A systematic review on diagnostic accuracy of CT-based detection of significant coronary artery disease. Eur J Radiol. 2008;65(3):449-61. </p>
      <p content-type="ref-to-delete">11. Raggi P, Callister TQ, Cooil B, He ZX, Lippolis NJ, Russo DJ, et al. Identification of patients at increased risk of first unheralded acute myocardial infarction by electron-beam computed tomography. Circulation.2000;101(8):850-5. </p>
      <p content-type="ref-to-delete">12. Guimaraes JA, Victor EG, de Britto Leite MR, Gomes JM, Victor Filho E, Reyes Liveras J. Reliability of the interpretation of coronary angiography by the simple visual method. Arq Bras Cardiol. 2000;74(4):300-8. </p>
      <p content-type="ref-to-delete">13. Banerjee S, Crook AM, Dawson JR, Timmis AD, Hemingway H. Magnitude and Consequences of Error in Coronary Angiography Interpretation (The ACRE Study). Am J Cardiol. 2000;85(3):309-4   </p>
      <p>
        <fn id="end-body2" fn-type="corresp">
          <label>
            <bold> Mailing Address:</bold>
          </label>
          <p> Marcelo Souza Hadlich </p>
          <p> Rua Marqu&#234;s de S&#227;o Vicente, 182/102, G&#225;vea</p>
          <p> Postal Code 22451040, Rio de Janeiro, RJ - Brasil</p>
          <p> E-mail: </p>
          <p><email xlink:href="marcelohadlich@cardiol.br">marcelohadlich@cardiol.br</email>, </p>
          <p>
            <email xlink:href="marcelohadlich@gmail.com">marcelohadlich@gmail.com</email>
          </p>
        </fn>
      </p>
      <p>Manuscript received November 4, 2011; manuscript revised November 4, 2011; accepted June 13, 2012.</p>
    </body>
  </sub-article>
</article>
