{"id":24314,"date":"2022-03-08T09:27:14","date_gmt":"2022-03-08T15:27:14","guid":{"rendered":"https:\/\/saluddigital.com\/?p=24314"},"modified":"2025-10-21T10:14:31","modified_gmt":"2025-10-21T16:14:31","slug":"mit-desarrolla-tecnicas-para-promover-la-equidad-en-modelos-de-aprendizaje-automatico-en-salud","status":"publish","type":"post","link":"https:\/\/saluddigital.com\/en\/big-data\/mit-desarrolla-tecnicas-para-promover-la-equidad-en-modelos-de-aprendizaje-automatico-en-salud\/","title":{"rendered":"MIT develops techniques to promote equity in machine learning models in health"},"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

The Massachusetts Institute of Technology (MIT) has developed a new technique to reduce bias and increase the range of machine learning models.<\/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

Artificial Intelligence, machine learning and deep learning models are key aspects for the development of research and studies that require the processing of large amounts of data. However, an imbalance in the data can lead to the creation of models that introduce bias into the research. For this reason, MIT has published a study showing how it was possible to increase fairness in machine learning models.<\/p>

The article titled: \u201cIs equity just deep metric? Assessing and addressing subgroup gaps in DML\u201d, explains that these models can be corrected. In this way they developed a technique that allows the model to produce fair results regardless of whether it was trained with unbalanced data.<\/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

\u201cIn machine learning, it is common to blame data for bias in models. But we don't always have balanced data. So we need to come up with methods that actually fix the problem with unbalanced data,\u201d says lead author Natalie Dulero, a graduate student in the Healthy Machine Learning Group at MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL).<\/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

\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\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 this sense, the correction of the models can adapt to new data and learn to group new types of information. \u201cWe know that data reflects the biases of processes in society. This means that we have to change our approach to design methods that better fit reality,\u201d explains lead author Marzyeh Ghassemi.<\/p>

Thanks to this type of development it is possible to improve models that have been successful for research. In the case of health, it is important to maintain high standards in deep machine learning models, or any other algorithm, that involves patients.<\/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

EUREKALERT<\/strong><\/p>

https:\/\/www.eurekalert.org\/news-releases\/945050<\/a><\/p>

HEALTH IT ANALYTICS<\/strong><\/p>

https:\/\/healthitanalytics.com\/news\/new-mit-technique-aims-to-boost-fairness-within-machine-learning-models<\/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":"

El Massachusetts Institute of Technology (MIT), ha desarrollado una nueva t\u00e9cnica para reducir el sesgo y aumentar el alcance de modelos de aprendizaje autom\u00e1tico. La Inteligencia Artificial, los modelos de aprendizaje autom\u00e1tico y aprendizaje profundo, son aspectos clave para el desarrollo de investigaciones y estudios que requieren el procesamiento de grandes cantidades de datos. Sin embargo, un desequilibrio en los datos puede derivar en la creaci\u00f3n de modelos que dan entrada a sesgos en la investigaci\u00f3n. Por ello el MIT ha publicado un estudio donde muestra c\u00f3mo fue posible aumentar la equidad en modelos de aprendizaje autom\u00e1tico. El art\u00edculo titulado: “\u00bfEs la equidad solo m\u00e9trica profunda? Evaluaci\u00f3n y abordaje de las brechas de subgrupos en DML”, explica que estos modelos pueden corregirse. De esta forma desarrollaron una t\u00e9cnica que permite que el modelo produzca resultados justos sin importar si este fue entrenado con datos desequilibrados. \u201cEn el aprendizaje autom\u00e1tico, es com\u00fan culpar a los datos por el sesgo en los modelos. Pero no siempre tenemos datos equilibrados. Por lo tanto, debemos idear m\u00e9todos que realmente solucionen el problema con datos desequilibrados\u201d, dice la autora principal Natalie Dulero, estudiante de posgrado en el Grupo de Aprendizaje Autom\u00e1tico Saludable del Laboratorio de Ciencias de la Computaci\u00f3n e Inteligencia Artificial (CSAIL) del MIT. En este sentido la correcci\u00f3n de los modelos puede adaptarse a nuevos datos y aprender a agrupar nuevos tipos de informaci\u00f3n. \u201cSabemos que los datos reflejan los sesgos de los procesos en la sociedad. Esto significa que tenemos que cambiar nuestro enfoque para dise\u00f1ar m\u00e9todos que se adapten mejor a la realidad\u201d, explica la autora principal Marzyeh Ghassemi. Gracias a este tipo de desarrollos es posible mejorar modelos que han sido exitosos para la investigaci\u00f3n. En el caso de la salud, es importante mantener est\u00e1ndares altos en los modelos de aprendizaje autom\u00e1tico profundo o cualquier otro algoritmo, que involucre a pacientes. BIBLIOGRAF\u00cdA EUREKALERT https:\/\/www.eurekalert.org\/news-releases\/945050 HEALTH IT ANALYTICS https:\/\/healthitanalytics.com\/news\/new-mit-technique-aims-to-boost-fairness-within-machine-learning-models<\/p>","protected":false},"author":1,"featured_media":24316,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[156,3393,160],"tags":[145],"class_list":["post-24314","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-big-data","category-inteligencia-artificial-y-ciencia","category-noticias","tag-noticias"],"aioseo_notices":[],"_links":{"self":[{"href":"https:\/\/saluddigital.com\/en\/wp-json\/wp\/v2\/posts\/24314","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=24314"}],"version-history":[{"count":0,"href":"https:\/\/saluddigital.com\/en\/wp-json\/wp\/v2\/posts\/24314\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/saluddigital.com\/en\/wp-json\/wp\/v2\/media\/24316"}],"wp:attachment":[{"href":"https:\/\/saluddigital.com\/en\/wp-json\/wp\/v2\/media?parent=24314"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/saluddigital.com\/en\/wp-json\/wp\/v2\/categories?post=24314"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/saluddigital.com\/en\/wp-json\/wp\/v2\/tags?post=24314"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}