A clinical deep learning framework for continually learning from cardiac signals across diseases, time, modalities, and institutions

Deep learning algorithms trained on data streamed temporally from different clinical sites and from a multitude of physiological sensors are generally affected by a degradation in performance. To mitigate this, the authors propose a continual learning strategy that employs a replay buffer.

Guardado en:

Detalles Bibliográficos
Autores principales:	Dani Kiyasseh, Tingting Zhu, David Clifton
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2021
Materias:	Science Q
Acceso en línea:	https://doaj.org/article/5943e24bee1a4eb1be3eaa04425d47c2
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:5943e24bee1a4eb1be3eaa04425d47c2
record_format	dspace
spelling	oai:doaj.org-article:5943e24bee1a4eb1be3eaa04425d47c22021-12-02T18:34:21ZA clinical deep learning framework for continually learning from cardiac signals across diseases, time, modalities, and institutions10.1038/s41467-021-24483-02041-1723https://doaj.org/article/5943e24bee1a4eb1be3eaa04425d47c22021-07-01T00:00:00Zhttps://doi.org/10.1038/s41467-021-24483-0https://doaj.org/toc/2041-1723Deep learning algorithms trained on data streamed temporally from different clinical sites and from a multitude of physiological sensors are generally affected by a degradation in performance. To mitigate this, the authors propose a continual learning strategy that employs a replay buffer.Dani KiyassehTingting ZhuDavid CliftonNature PortfolioarticleScienceQENNature Communications, Vol 12, Iss 1, Pp 1-11 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Science Q
spellingShingle	Science Q Dani Kiyasseh Tingting Zhu David Clifton A clinical deep learning framework for continually learning from cardiac signals across diseases, time, modalities, and institutions
description	Deep learning algorithms trained on data streamed temporally from different clinical sites and from a multitude of physiological sensors are generally affected by a degradation in performance. To mitigate this, the authors propose a continual learning strategy that employs a replay buffer.
format	article
author	Dani Kiyasseh Tingting Zhu David Clifton
author_facet	Dani Kiyasseh Tingting Zhu David Clifton
author_sort	Dani Kiyasseh
title	A clinical deep learning framework for continually learning from cardiac signals across diseases, time, modalities, and institutions
title_short	A clinical deep learning framework for continually learning from cardiac signals across diseases, time, modalities, and institutions
title_full	A clinical deep learning framework for continually learning from cardiac signals across diseases, time, modalities, and institutions
title_fullStr	A clinical deep learning framework for continually learning from cardiac signals across diseases, time, modalities, and institutions
title_full_unstemmed	A clinical deep learning framework for continually learning from cardiac signals across diseases, time, modalities, and institutions
title_sort	clinical deep learning framework for continually learning from cardiac signals across diseases, time, modalities, and institutions
publisher	Nature Portfolio
publishDate	2021
url	https://doaj.org/article/5943e24bee1a4eb1be3eaa04425d47c2
work_keys_str_mv	AT danikiyasseh aclinicaldeeplearningframeworkforcontinuallylearningfromcardiacsignalsacrossdiseasestimemodalitiesandinstitutions AT tingtingzhu aclinicaldeeplearningframeworkforcontinuallylearningfromcardiacsignalsacrossdiseasestimemodalitiesandinstitutions AT davidclifton aclinicaldeeplearningframeworkforcontinuallylearningfromcardiacsignalsacrossdiseasestimemodalitiesandinstitutions AT danikiyasseh clinicaldeeplearningframeworkforcontinuallylearningfromcardiacsignalsacrossdiseasestimemodalitiesandinstitutions AT tingtingzhu clinicaldeeplearningframeworkforcontinuallylearningfromcardiacsignalsacrossdiseasestimemodalitiesandinstitutions AT davidclifton clinicaldeeplearningframeworkforcontinuallylearningfromcardiacsignalsacrossdiseasestimemodalitiesandinstitutions
_version_	1718377861375590400

A clinical deep learning framework for continually learning from cardiac signals across diseases, time, modalities, and institutions

Ejemplares similares