Embracing cohort heterogeneity in clinical machine learning development: a step toward generalizable models

Autor: Michiel Schinkel, Frank C. Bennis, Anneroos W. Boerman, W. Joost Wiersinga, Prabath W. B. Nanayakkara
Přispěvatelé: Internal medicine, ACS - Diabetes & metabolism, APH - Digital Health, APH - Quality of Care, Center of Experimental and Molecular Medicine, Graduate School, Paediatrics, Infectious diseases, AII - Cancer immunology, AII - Infectious diseases, APH - Global Health
Jazyk: angličtina
Rok vydání: 2023
Předmět:
Zdroj: Scientific Reports, 13(1):8363. Nature Publishing Group
Schinkel, M, Bennis, F C, Boerman, A W, Wiersinga, W J & Nanayakkara, P W B 2023, ' Embracing cohort heterogeneity in clinical machine learning development : a step toward generalizable models ', Scientific Reports, vol. 13, no. 1, 8363 . https://doi.org/10.1038/s41598-023-35557-y
Scientific reports, 13(1):8363. Nature Publishing Group
ISSN: 2045-2322
Popis: This study is a simple illustration of the benefit of averaging over cohorts, rather than developing a prediction model from a single cohort. We show that models trained on data from multiple cohorts can perform significantly better in new settings than models based on the same amount of training data but from just a single cohort. Although this concept seems simple and obvious, no current prediction model development guidelines recommend such an approach.
Databáze: OpenAIRE
Externí odkaz: