{"id":12646,"date":"2021-03-01T09:14:19","date_gmt":"2021-03-01T15:14:19","guid":{"rendered":"https:\/\/saluddigital.com\/?p=12646"},"modified":"2025-10-21T13:00:43","modified_gmt":"2025-10-21T19:00:43","slug":"modelo-de-aprendizaje-profundo-diagnostica-covid-19-a-traves-de-tomografias-computarizadas","status":"publish","type":"post","link":"https:\/\/saluddigital.com\/en\/avance-de-la-ciencia\/modelo-de-aprendizaje-profundo-diagnostica-covid-19-a-traves-de-tomografias-computarizadas\/","title":{"rendered":"Deep learning model diagnoses COVID-19 through CT scans"},"content":{"rendered":"
\n\t\t\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t

CovidCTNet, is an open-source deep learning-based platform that is capable of diagnosing COVID-19 using small CT scan images. <\/h2>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

Timely and accurate diagnosis of COVID-19 is crucial to reduce not only the spread of the disease but also mortality. The most widely used conventional tests in the world for the detection of COVID-19 are polymerase chain reaction or PCR, one of the most reliable tests. However, its detection accuracy reaches 70 to 75%. Another way to know or make a diagnosis of a person for COVID-19, is by computed tomography (CT) images, these images have a higher sensitivity of almost 98%, however, the accuracy is 70%.<\/p>

Medical researchers from universities in Iran, the United States, and Vietnam developed \u201can open-source framework, CovidCTNet, composed of a set of deep learning algorithms that accurately differentiates Covid-19 from community-acquired pneumonia (CAP) and other lung diseases.\u201d This framework increases accuracy markedly in CT image detection, reaching 95%, far surpassing traditional radiology work that achieves 70% accuracy.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t

\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\"\"\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

\u201cCovidCTNet is designed to work with heterogeneous and small sample sizes independent of the CT imaging hardware.\u201d This innovation aims to facilitate the detection of COVID-19 worldwide and also facilitate the work of radiologists in this process, which is why the developed algorithms as well as the model parameters are available as an open-source model.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t

\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

\u201cOpen-source sharing of CovidCTNet enables developers to rapidly improve and optimize services while preserving user privacy and data ownership,\u201d the authors explain in the research published in Nature.<\/p>

In its results CovidCTNet, evaluated a data set of 20 mixed control cases, first using the developed algorithm and simultaneously by four independent radiologists. \" The average reader performance of four radiologists showed a sensitivity of 79% for Covid-19 and specificity of 82.14%. The CNN classification of CovidCTNet, however outperformed the radiologists and achieved Covid-19 detection with sensitivity and specificity of 93 and 100%, respectively details the comparison of radiologist performance versus CovidCTNet,\u201d the study explains.<\/p>

Nevertheless, this technology is not intended to replace the work of radiologists, but rather to streamline their work, and carry out better diagnoses.\u00a0<\/p>

\u00a0<\/p>

The complete research is freely available in the journal Nature https:\/\/www.nature.com\/articles\/s41746-021-00399-3<\/a><\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t

\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/i><\/span>\n\t\t\t\t\t\t\t\t<\/i><\/span>\n\t\t\t\t\t\t\t\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\t\t\t BIBLIOGRAPHY<\/a>\n\t\t\t\t\t<\/div>\n\n\t\t\t\t\t

NATURE <\/strong><\/p>

https:\/\/www.nature.com\/articles\/s41746-021-00399-3<\/a><\/p>

\u00a0<\/strong><\/p>

THE LANCET \u00a0<\/strong><\/p>

https:\/\/www.thelancet.com\/journals\/landig\/article\/PIIS2589-7500(20)30142-4\/fulltext#tbl1<\/a><\/p><\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<\/div>","protected":false},"excerpt":{"rendered":"

CovidCTNet, es una plataforma de c\u00f3digo abierto basada en aprendizaje profundo que es capaz de diagnosticar COVID-19 utilizando peque\u00f1as im\u00e1genes de tomograf\u00edas computarizadas. El diagn\u00f3stico oportuno y preciso de COVID-19 es crucial para reducir no solo la propagaci\u00f3n de la enfermedad sino la mortalidad. Las pruebas convencionales m\u00e1s utilizadas en el mundo para la detecci\u00f3n de COVID-19 son las de detecci\u00f3n por reacci\u00f3n en cadena de la polimerasa o PCR, es una de las pruebas m\u00e1s confiables. Sin embargo, su precisi\u00f3n de detecci\u00f3n alcanza un 70 a75%. Otra de las formas de conocer o realizar un diagn\u00f3stico de una persona por COVID-19, es mediante im\u00e1genes de tomograf\u00eda computarizada (TC), estas im\u00e1genes tienen una mayor sensibilidad de casi de 98%, sin embargo, la precisi\u00f3n es del 70%. Investigadores m\u00e9dicos de universidades de Ir\u00e1n, Estados Unidos y Vietnam desarrollaron “un marco de c\u00f3digo abierto, CovidCTNet, compuesto por un conjunto de algoritmos de aprendizaje profundo que diferencia con precisi\u00f3n Covid-19 de la neumon\u00eda adquirida en la comunidad (NAC) y otras enfermedades pulmonares”. Este marco aumenta la precisi\u00f3n notablemente en la detecci\u00f3n de im\u00e1genes por TC, llegando a un 95%, superando con creces a trabajos de radiolog\u00eda tradicional que logran precisi\u00f3n de 70%. “CovidCTNet est\u00e1 dise\u00f1ado para trabajar con tama\u00f1os de muestra peque\u00f1os y heterog\u00e9neos, independientemente del hardware de im\u00e1genes de TC”. Esta innovaci\u00f3n tiene como objetivo facilitar la detecci\u00f3n de COVID-19 en todo el mundo y adem\u00e1s facilitar el trabajo de los radi\u00f3logos en este proceso, es por ello que los algoritmos desarrollados, as\u00ed como los par\u00e1metros del modelo est\u00e1n disponible como un modelo de c\u00f3digo abierto. “El uso compartido de c\u00f3digo abierto de CovidCTNet permite a los desarrolladores mejorar y optimizar r\u00e1pidamente los servicios mientras preservan la privacidad del usuario y la propiedad de los datos. Estamos lanzando todos los algoritmos y detalles de los par\u00e1metros del modelo como c\u00f3digo abierto”, explican los autores en la investigaci\u00f3n publicada en Nature. En sus resultados CovidCTNet, evalu\u00f3 un conjunto de datos de 20 caso mixtos de control, primero utilizando el algoritmo desarrollado y simult\u00e1neamente por cuatro radi\u00f3logos independientes. \u201cEl rendimiento medio del lector de cuatro radi\u00f3logos mostr\u00f3 una sensibilidad del 79% para Covid-19 y una especificidad del 82,14%. Sin embargo, la clasificaci\u00f3n de CNN de CovidCTNet super\u00f3 a los radi\u00f3logos y logr\u00f3 la detecci\u00f3n de Covid-19 con una sensibilidad y especificidad del 93 y 100%, respectivamente\u201d, explica el estudio. Sin embargo, esta tecnolog\u00eda no intenta remplazar el trabajo de los radi\u00f3logos sino agilizar su trabajo, y llevar a cabo mejores diagn\u00f3sticos.\u00a0 \u00a0 La investigaci\u00f3n completa est\u00e1 disponible de manera libre en la revista Nature https:\/\/www.nature.com\/articles\/s41746-021-00399-3 BIBLIOGRAF\u00cdA NATURE https:\/\/www.nature.com\/articles\/s41746-021-00399-3 \u00a0 THE LANCET \u00a0 https:\/\/www.thelancet.com\/journals\/landig\/article\/PIIS2589-7500(20)30142-4\/fulltext#tbl1<\/p>","protected":false},"author":1,"featured_media":12647,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[152,156],"tags":[145],"class_list":["post-12646","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-avance-de-la-ciencia","category-big-data","tag-noticias"],"aioseo_notices":[],"_links":{"self":[{"href":"https:\/\/saluddigital.com\/en\/wp-json\/wp\/v2\/posts\/12646","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/saluddigital.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/saluddigital.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/saluddigital.com\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/saluddigital.com\/en\/wp-json\/wp\/v2\/comments?post=12646"}],"version-history":[{"count":0,"href":"https:\/\/saluddigital.com\/en\/wp-json\/wp\/v2\/posts\/12646\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/saluddigital.com\/en\/wp-json\/wp\/v2\/media\/12647"}],"wp:attachment":[{"href":"https:\/\/saluddigital.com\/en\/wp-json\/wp\/v2\/media?parent=12646"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/saluddigital.com\/en\/wp-json\/wp\/v2\/categories?post=12646"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/saluddigital.com\/en\/wp-json\/wp\/v2\/tags?post=12646"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}