{"id":13627,"date":"2021-04-16T09:42:30","date_gmt":"2021-04-16T14:42:30","guid":{"rendered":"https:\/\/saluddigital.com\/?p=13627"},"modified":"2025-10-21T12:49:15","modified_gmt":"2025-10-21T18:49:15","slug":"recomendaciones-en-la-utilizacion-de-inteligencia-artificial-para-la-deteccion-de-covid-19","status":"publish","type":"post","link":"https:\/\/saluddigital.com\/en\/big-data\/recomendaciones-en-la-utilizacion-de-inteligencia-artificial-para-la-deteccion-de-covid-19\/","title":{"rendered":"Recommendations for using Artificial Intelligence for COVID-19 detection"},"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

Thanks to machine and deep learning methods it is possible to perform COVID-19 detections and prognoses, using chest X-rays and CT scans. However, despite advances in the last year, no model is suitable for clinical use yet.<\/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

In addition to PCR testing, specialists have supplemented that information with chest X-rays, for a more complete COVID-19 diagnosis. In addition, in some countries where sufficient PCR testing is not available, imaging could be a substitute.<\/p>

Since the beginning of the pandemic, researchers from different countries started to develop Artificial Intelligence (AI) models, based on machine learning, to improve the diagnosis of COVID-19. According to an analysis published in Nature Machine Intelligence<\/em>, which analyzed more than 300 studies that used deep learning-based and 69 machine learning-based models, the authors explained that, \u201cDespite the huge efforts of researchers to develop machine learning models for COVID-19 diagnosis and prognosis, we found methodological flaws and many biases throughout the literature, leading to highly optimistic reported performance.\u201d<\/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

\u201cAny machine learning algorithm is only as good as the data it\u2019s trained on. Especially for a brand-new disease like COVID-19, it's vital that the training data is as diverse as possible because, as we've seen throughout this pandemic, there are many different factors that affect what the disease looks like and how it behaves\u201d, explained Dr. Michael Roberts, Department of Applied Mathematics and Theoretical Physics, University of Cambridge.<\/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

In the end, only 62 of the studies were analyzed as they met the criteria set by the authors. However, as the authors explain \"in their current reported form, none of the machine learning models included in this review are likely candidates for clinical translation for the diagnosis\/prognosis of COVID-19.\"<\/p>

The authors made five main recommendations for improving COVID-19 detection and diagnostic models: (1) considerations when collating COVID-19 imaging datasets that are to be made public; (2) methodological considerations for algorithm developers; (3) specific issues about reproducibility of the results in the literature; (4) considerations for authors to ensure sufficient documentation of methodologies in manuscripts; and (5) considerations for reviewers performing peer review of manuscripts.<\/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\/s42256-021-00307-0<\/a><\/p>

\u00a0<\/p>

RADIOLOGY BUSINESS <\/strong><\/p>

https:\/\/www.radiologybusiness.com\/topics\/artificial-intelligence\/imaging-ai-models-covid-19-diagnosis-clinical-use<\/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":"

Gracias a m\u00e9todos de aprendizaje autom\u00e1tico y profundo es posible realizar detecciones y pron\u00f3sticos de COVID-19, utilizando radiograf\u00edas y tomograf\u00edas computarizadas de t\u00f3rax. Sin embargo, a pesar de los avances del \u00faltimo a\u00f1o ning\u00fan modelo es apto para uso cl\u00ednico a\u00fan. Adem\u00e1s de las pruebas PCR, los especialistas han complementado esa informaci\u00f3n con radiograf\u00edas de t\u00f3rax, para un diagn\u00f3stico de COVID-19 m\u00e1s completo. Adem\u00e1s, en algunos pa\u00edses donde no hay suficientes pruebas PCR disponibles, el diagn\u00f3stico por im\u00e1genes podr\u00eda ser un substituto. Desde el comienzo de la pandemia, investigadores de diferentes pa\u00edses comenzaron a desarrollar modelos de Inteligencia Artificial (IA), basados en aprendizaje autom\u00e1tico, para mejorar el diagnostico de COVID-19. Seg\u00fan un an\u00e1lisis publicado en Nature Machine Intelligence, que analiz\u00f3 m\u00e1s de 300 estudios que utilizaron modelos basados en aprendizaje profundo y 69 basados en aprendizaje autom\u00e1tico, los autores explicaron que: \u201cA pesar de los enormes esfuerzos de los investigadores para desarrollar modelos de aprendizaje autom\u00e1tico para el diagn\u00f3stico y pron\u00f3stico de COVID-19, encontramos fallas metodol\u00f3gicas y muchos sesgos a lo largo de la literatura, lo que llev\u00f3 a un rendimiento informado altamente optimista\u201d. \u201cCualquier algoritmo de aprendizaje autom\u00e1tico es tan bueno como los datos con los que est\u00e1 entrenado. Especialmente para una enfermedad nueva como COVID-19, es vital que los datos de entrenamiento sean lo m\u00e1s diversos posible porque, como hemos visto a lo largo de esta pandemia, hay muchos factores diferentes que afectan el aspecto de la enfermedad y c\u00f3mo se manifiesta\u201d, explic\u00f3 el Dr. Michael Roberts, del Departamento de Matem\u00e1ticas Aplicadas y F\u00edsica Te\u00f3rica de la Universidad de Cambridge. Finalmente, solo 62 de los estudios fueron analizados al cumplir con los criterios establecidos por los autores. Sin embargo, seg\u00fan explican los autores \u201cen su forma informada actual, ninguno de los modelos de aprendizaje autom\u00e1tico incluidos en esta revisi\u00f3n son candidatos probables para la traducci\u00f3n cl\u00ednica para el diagn\u00f3stico \/ pron\u00f3stico de COVID-19\u201d. Los autores realizaron cinco recomendaciones principales para mejorar los modelos de detecci\u00f3n  y diagn\u00f3stico  de COVID-19: (1) consideraciones al recopilar conjuntos de datos de im\u00e1genes de COVID-19 que se har\u00e1n p\u00fablicos; (2) consideraciones metodol\u00f3gicas para los desarrolladores de algoritmos; (3) cuestiones espec\u00edficas sobre la reproducibilidad de los resultados en la literatura; (4) consideraciones para que los autores aseguren una documentaci\u00f3n suficiente de las metodolog\u00edas en los manuscritos; y (5) consideraciones para los revisores que realizan la revisi\u00f3n por pares de los manuscritos. BIBLIOGRAF\u00cdA NATURE https:\/\/www.nature.com\/articles\/s42256-021-00307-0 RADIOLOGY BUSINESS https:\/\/www.radiologybusiness.com\/topics\/artificial-intelligence\/imaging-ai-models-covid-19-diagnosis-clinical-use<\/p>","protected":false},"author":1,"featured_media":13628,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[3399,156,3393,1418],"tags":[145],"class_list":["post-13627","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ai-analitica","category-big-data","category-inteligencia-artificial-y-ciencia","category-resena-de-publicaciones-cientificas","tag-noticias"],"aioseo_notices":[],"_links":{"self":[{"href":"https:\/\/saluddigital.com\/en\/wp-json\/wp\/v2\/posts\/13627","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=13627"}],"version-history":[{"count":0,"href":"https:\/\/saluddigital.com\/en\/wp-json\/wp\/v2\/posts\/13627\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/saluddigital.com\/en\/wp-json\/wp\/v2\/media\/13628"}],"wp:attachment":[{"href":"https:\/\/saluddigital.com\/en\/wp-json\/wp\/v2\/media?parent=13627"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/saluddigital.com\/en\/wp-json\/wp\/v2\/categories?post=13627"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/saluddigital.com\/en\/wp-json\/wp\/v2\/tags?post=13627"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}