Identificación no supervisada de zona más relevante en radiografía para la detección de COVID
Cargando...
Archivos
Fecha
2021
Autores
Profesor/a Guía
Facultad/escuela
Idioma
es
Título de la revista
ISSN de la revista
Título del volumen
Editor
Universidad Andrés Bello
Nombre de Curso
Licencia CC
Licencia CC
Resumen
Desde principios del año 2020 el diagnóstico del virus COVID-19 es un problema de
vital importancia que ha afectado la vida de millones de personas a lo largo de todo el
mundo debido a que, en muchos casos, este se realiza con mucha lentitud, con un tiempo
estimado que varía entre 1 y 5 días.
Una forma eficiente y rápida de detectar la presencia, tanto del virus COVID-19 como
también la presencia de otras enfermedades, es utilizando Inteligencia Artificial aplicada
a las imágenes obtenidas en la toma de radiografía de pulmones. Actualmente, esta detección IA de COVID-19 considera la imagen completa, pero no detecta la parte que más aporta a la presencia de esta, la cual podría variar entre personas.
Nosotros proponemos usar un algoritmo basado en aprendizaje profundo para aprender a reconocer la zona más relevante para detectar COVID-19 basadas en regiones de imagen y sin supervisión humana.
Since the beginning of 2020, the diagnostic of the COVID-19 virus has been a problem of vitally importance that has affected the lives of millions of people throughout the world because, in many cases, it is done very slowly, with an estimated time that varies between 1 and 5 days. An efficient and fastest way to detect the presence of both the COVID-19 virus as well as the presence of other diseases is by using Artificial Intelligence applied to the images obtained from taking a lung X-ray. Currently, this AI detection of COVID-19 considers the complete image, but does not detect the part that contributes the most to its presence, which could vary between people.
Since the beginning of 2020, the diagnostic of the COVID-19 virus has been a problem of vitally importance that has affected the lives of millions of people throughout the world because, in many cases, it is done very slowly, with an estimated time that varies between 1 and 5 days. An efficient and fastest way to detect the presence of both the COVID-19 virus as well as the presence of other diseases is by using Artificial Intelligence applied to the images obtained from taking a lung X-ray. Currently, this AI detection of COVID-19 considers the complete image, but does not detect the part that contributes the most to its presence, which could vary between people.
Notas
Tesis (Ingeniero Civil Informático)
Palabras clave
Inteligencia Artificial, Redes Neurales (Ciencia de la Computación), COVID-19 (Enfermedad)