Challenges of Estimating Global Feature Importance in Real-World Health Care Data.
Stud Health Technol Inform
; 302: 1057-1061, 2023 May 18.
Article
em En
| MEDLINE
| ID: mdl-37203580
ABSTRACT
Feature importance is often used to explain clinical prediction models. In this work, we examine three challenges using experiments with electronic health record data computational feasibility, choosing between methods, and interpretation of the resulting explanation. This work aims to create awareness of the disagreement between feature importance methods and underscores the need for guidance to practitioners how to deal with these discrepancies.
Palavras-chave
Texto completo:
1
Base de dados:
MEDLINE
Assunto principal:
Saúde Global
/
Registros Eletrônicos de Saúde
Tipo de estudo:
Guideline
/
Prognostic_studies
Idioma:
En
Ano de publicação:
2023
Tipo de documento:
Article