Subpopulation-specific machine learning prognosis for underrepresented patients with double prioritized bias correction

BACKGROUND: Many clinical datasets are intrinsically imbalanced, dominated by overwhelming majority groups. Off-the-shelf machine learning models that optimize the prognosis of majority patient types (e.g., healthy class) may cause substantial errors on the minority prediction class (e.g., disease c...

Descripción completa

Detalles Bibliográficos
Autores principales: Afrose, Sharmin, Song, Wenjia, Nemeroff, Charles B., Lu, Chang, Yao, Danfeng (Daphne)
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Nature Publishing Group UK 2022
Materias:
Article
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9436942/
https://www.ncbi.nlm.nih.gov/pubmed/36059892
http://dx.doi.org/10.1038/s43856-022-00165-w

Internet

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9436942/
https://www.ncbi.nlm.nih.gov/pubmed/36059892
http://dx.doi.org/10.1038/s43856-022-00165-w

Ejemplares similares