Mind the Gap — Machine Learning, Dataset Shift, and History in the Age of Clinical Algorithms

1 min read Original article ↗

Published January 20, 2024

N Engl J Med 2024;390:293-295

DOI: 10.1056/NEJMp2311015

Audio Interview

Abstract

In the 1970s, developers of an early computerized diagnostic program encountered the problem of “dataset shift” — a challenge that is relevant today as machine-learning algorithms spread throughout clinical practice.

Notes

This article was published on January 20, 2024, at NEJM.org.

Disclosure forms provided by the authors are available at NEJM.org.

Supplementary Material

Disclosure Forms (nejmp2311015_disclosures.pdf)