{"id":13342,"date":"2021-04-07T09:00:00","date_gmt":"2021-04-07T14:00:00","guid":{"rendered":"https:\/\/saluddigital.com\/?p=13342"},"modified":"2025-10-21T12:51:58","modified_gmt":"2025-10-21T18:51:58","slug":"algoritmo-de-aprendizaje-profundo-detecta-lesiones-pigmentadas-para-el-diagnostico-de-melanoma","status":"publish","type":"post","link":"https:\/\/saluddigital.com\/en\/big-data\/algoritmo-de-aprendizaje-profundo-detecta-lesiones-pigmentadas-para-el-diagnostico-de-melanoma\/","title":{"rendered":"Deep learning algorithm detects pigmented lesions for melanoma diagnosis"},"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

Dermatological researchers and specialists in engineering and Artificial Intelligence (AI), developed an algorithm through a convolutional neural network to examine skin lesions, through photographs, even mobile phone photos.<\/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 the United States in 2019 more than 96,000 people were diagnosed with melanoma, a disease, 7,230 people died that year. The process of early identification of suspicious pigmented lesions (SPL) is important to reduce the cost of treatment by up to 20 times. However, the number of efficient tools for SPL detection is still limited in a country like the United States. That is why the researchers developed this algorithm to improve diagnosis.<\/p>

The results were published in Science Translational Medicine<\/em> under the title: \u201cUsing deep learning for dermatologist-level detection of suspicious pigmented skin lesions from wide-field images.\u201d<\/p>

MIT, Harvard and other institutions researchers built a deep learning-based algorithm for evaluating skin lesions and classifying them to rule out diseases such as skin cancer. \u201cRather than evaluate a single lesion at a time looking for predetermined signs of neoplasia, the algorithm identifies lesions that differ from most of the other marks on that patient\u2019s skin, flagging them for further examination and ranking them in order of suspiciousness. The algorithm performed similarly to board-certified dermatologists and could potentially be used at primary care visits to help clinicians triage suspicious lesions for follow-up.\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

The algorithm's operation is based on deep convoluted neural networks DCNNs, which through analysis of wide-field photographs is able to obtain large samples of skin sections for diagnosing diseases before they develop.<\/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

38,283 dermatological data from 133 patients and other publicly available images were collected for the study. \u201cOur system achieved more than 90.3% sensitivity (95% confidence interval, 90 to 90.6) and 89.9% specificity (89.6 to 90.2%) in distinguishing SPLs from nonsuspicious lesions, skin, and complex backgrounds, avoiding the need for cumbersome individual lesion imaging,\u201d the study explains.<\/p>

MIT researchers contacted in 2014 the Spanish physician, Dr. Jos\u00e9 Avil\u00e9s Izquierdo, from the Hospital General Universitario Gregorio Ma\u00f1\u00f3n in Madrid, -who is one of the co-authors of the study published in the journal Science<\/em>- for the development of a machine learning<\/em>project for the detection of malignant melanomas. According to Dr. Avil\u00e9s Izquierdo, the purpose of the study was \"to develop an algorithm that would serve as an evaluation platform for different physical and computer tools aimed at the early diagnosis of skin cancer, especially melanoma.<\/p>

Dr. Avil\u00e9s Izquierdo reported his experience on the specialized site Medscape: \"The researchers at the Massachusetts Institute of Technology<\/em> developed a system for analyzing images of suspicious pigmented lesions based on 399 variables related to symmetry, borders, colors, texture or size. They analyzed 38,283 images from patients and public resources (image banks, atlases, and search engines)\".<\/p>

You can find the link to the study at the following link: https:\/\/stm.sciencemag.org\/content\/13\/581\/eabb3652<\/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

SCIENCE<\/strong><\/p>

https:\/\/stm.sciencemag.org\/content\/13\/581\/eabb3652<\/a><\/p>

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

MEDSCAPE<\/strong><\/p>

https:\/\/espanol.medscape.com\/verarticulo\/5906762?pa=yD%2BfwAM89jOM%2FCjJ48DFDxEl9jnXXTCHdaFdFnBfgo5dHhVxogEebHpXEwVpjYWg56MI7dGTgNawPfsOtJla9Q%3D%3D<\/a><\/p>

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

WITH HEALTH <\/strong><\/p>

https:\/\/www.consalud.es\/tecnologia\/algoritmo-inteligencia-artificial-lesiones-pigmentadas-precision_94618_102.html<\/a><\/p>

\u00a0<\/p>

MYPRESS MX<\/strong><\/p>

https:\/\/www.mypress.mx\/tecnologia\/mit-metodo-para-detectar-cancer-de-piel-inteligencia-artificial-9009<\/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":"

Investigadores dermatol\u00f3gicos y especialistas en ingenier\u00eda e Inteligencia Artificial (IA), desarrollaron un algoritmo a trav\u00e9s de una red neuronal convolucional para examinar lesiones en la piel, a trav\u00e9s de fotograf\u00edas, incluso tomadas utilizando un tel\u00e9fono m\u00f3vil. En Estados Unidos en 2019 m\u00e1s de 96 mil personas fueron diagnosticadas con melanoma, enfermedad que cost\u00f3 la vida de 7,230 personas. El proceso de identificaci\u00f3n temprana de lesiones pigmentadas sospechosas (SPL, en ingl\u00e9s) es importante para reducir el costo del tratamiento hasta 20 veces. Sin embargo, la cantidad de herramientas eficientes para la detecci\u00f3n de SPL, es limitada a\u00fan en un pa\u00eds como Estados Unidos. Es por ello que los investigadores desarrollaron este algoritmo para mejorar el diagn\u00f3stico. Los resultados fueron publicados en Science Translational Medicine bajo el t\u00edtulo: \u201cUso del aprendizaje profundo para la detecci\u00f3n a nivel dermatol\u00f3gico de lesiones cut\u00e1neas pigmentadas sospechosas a partir de im\u00e1genes de campo amplio\u201d. Investigadores del MIT, Harvard y otras instituciones construyeron un algoritmo basado en aprendizaje profundo para evaluar lesiones en la piel y clasific\u00e1ndolas para descartar enfermedades como c\u00e1ncer de piel. “En lugar de evaluar una sola lesi\u00f3n a la vez en busca de signos predeterminados de neoplasia, el algoritmo identifica las lesiones que difieren de la mayor\u00eda de las otras marcas en la piel de ese paciente, marc\u00e1ndolas para un examen m\u00e1s detallado y clasific\u00e1ndolas en orden de sospecha. El algoritmo funcion\u00f3 de manera similar a los dermat\u00f3logos certificados por la junta y podr\u00eda usarse potencialmente en las visitas de atenci\u00f3n primaria para ayudar a los m\u00e9dicos a clasificar las lesiones sospechosas para el seguimiento”. El funcionamiento del algoritmo est\u00e1 basado en redes neuronales profundas y enrevesadas DCNNs (por sus siglas en ingl\u00e9s), que a trav\u00e9s del an\u00e1lisis de fotograf\u00edas de amplio campo es posible obtener muestras grandes de secciones de la piel para el diagn\u00f3stico de enfermedades antes de que se desarrollen. Para el estudio fueron recopilados 38,283 datos dermatol\u00f3gicos de 133 pacientes y otras im\u00e1genes disponibles p\u00fablicamente. \u201cNuestro sistema logr\u00f3 m\u00e1s del 90,3% de sensibilidad (intervalo de confianza del 95%, 90 a 90,6) y 89,9% de especificidad (89,6 a 90,2%) en la distinci\u00f3n de SPL de lesiones no sospechosas, piel y fondos complejos, evitando la necesidad de im\u00e1genes de lesiones individuales engorrosas\u201d, explica el estudio. Investigadores del MIT contactaron en 2014 al m\u00e9dico espa\u00f1ol, Dr. Jos\u00e9 Avil\u00e9s Izquierdo, del Hospital General Universitario Gregorio Ma\u00f1\u00f3n de Madrid, -quien es uno de los coautores del estudio publicado en la revista Science, – para el desarrollo de un proyecto de aprendizaje autom\u00e1tico (o machine learning), para la detecci\u00f3n de melanomas malignos. Seg\u00fan el Dr. Avil\u00e9s Izquierdo, el prop\u00f3sito del estudio era \u201cdesarrollar un algoritmo que sirviera como plataforma de evaluaci\u00f3n para distintas herramientas f\u00edsicas e inform\u00e1ticas dirigidas al diagn\u00f3stico precoz del c\u00e1ncer de piel, especialmente del melanoma. El Dr. Avil\u00e9s Izquierdo, cont\u00f3 su experiencia en el sitio especializado Medscape: \u201cLos investigadores del Massachusetts Institute of Technology desarrollaron un sistema de an\u00e1lisis de im\u00e1genes de lesiones pigmentadas sospechosas basado en 399 variables relacionadas con simetr\u00eda, bordes, colores, textura o tama\u00f1o. Para ello analizaron 38,283 provenientes de pacientes y de recursos p\u00fablicos (bancos de im\u00e1genes, atlas y buscadores)\u201d. Puedes encontrar enlace del estudio en el siguiente enlace: https:\/\/stm.sciencemag.org\/content\/13\/581\/eabb3652 BIBLIOGRAF\u00cdA SCIENCE https:\/\/stm.sciencemag.org\/content\/13\/581\/eabb3652   MEDSCAPE https:\/\/espanol.medscape.com\/verarticulo\/5906762?pa=yD%2BfwAM89jOM%2FCjJ48DFDxEl9jnXXTCHdaFdFnBfgo5dHhVxogEebHpXEwVpjYWg56MI7dGTgNawPfsOtJla9Q%3D%3D   CONSALUD https:\/\/www.consalud.es\/tecnologia\/algoritmo-inteligencia-artificial-lesiones-pigmentadas-precision_94618_102.html MYPRESS MX https:\/\/www.mypress.mx\/tecnologia\/mit-metodo-para-detectar-cancer-de-piel-inteligencia-artificial-9009<\/p>","protected":false},"author":1,"featured_media":13343,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[3399,156,3393,1418],"tags":[145],"class_list":["post-13342","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\/13342","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=13342"}],"version-history":[{"count":0,"href":"https:\/\/saluddigital.com\/en\/wp-json\/wp\/v2\/posts\/13342\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/saluddigital.com\/en\/wp-json\/wp\/v2\/media\/13343"}],"wp:attachment":[{"href":"https:\/\/saluddigital.com\/en\/wp-json\/wp\/v2\/media?parent=13342"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/saluddigital.com\/en\/wp-json\/wp\/v2\/categories?post=13342"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/saluddigital.com\/en\/wp-json\/wp\/v2\/tags?post=13342"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}