Reproducibility in machine learning for health research: Still a ways to go

Matthew B.A. McDermott, Shirly Wang, Nikki Marinsek, Rajesh Ranganath, Luca Foschini, Marzyeh Ghassemi

Research output: Contribution to journalReview articlepeer-review

Abstract

Machine learning for health must be reproducible to ensure reliable clinical use. We evaluated 511 scientific papers across several machine learning subfields and found that machine learning for health compared poorly to other areas regarding reproducibility metrics, such as dataset and code accessibility. We propose recommendations to address this problem.

Original languageEnglish (US)
Article numbereabb1655
JournalScience Translational Medicine
Volume13
Issue number586
DOIs
StatePublished - Mar 24 2021

ASJC Scopus subject areas

  • Medicine(all)

Fingerprint Dive into the research topics of 'Reproducibility in machine learning for health research: Still a ways to go'. Together they form a unique fingerprint.

Cite this