Skip to Main content Skip to Navigation
Journal articles

Preventing dataset shift from breaking machine-learning biomarkers

Abstract : Machine learning brings the hope of finding new biomarkers extracted from cohorts with rich biomedical measurements. A good biomarker is one that gives reliable detection of the corresponding condition. However, biomarkers are often extracted from a cohort that differs from the target population. Such a mismatch, known as a dataset shift, can undermine the application of the biomarker to new individuals. Dataset shifts are frequent in biomedical research, e.g. because of recruitment biases. When a dataset shift occurs, standard machine-learning techniques do not suffice to extract and validate biomarkers. This article provides an overview of when and how dataset shifts breaks machine-learning extracted biomarkers, as well as detection and correction strategies.
Complete list of metadata
Contributor : Jérôme Dockès Connect in order to contact the contributor
Submitted on : Tuesday, July 20, 2021 - 11:20:16 PM
Last modification on : Friday, July 23, 2021 - 3:47:49 AM
Long-term archiving on: : Thursday, October 21, 2021 - 7:06:44 PM


Files produced by the author(s)


  • HAL Id : hal-03293375, version 1
  • ARXIV : 2107.09947


Jérôme Dockès, Gaël Varoquaux, Jean-Baptiste Poline. Preventing dataset shift from breaking machine-learning biomarkers. GigaScience, BioMed Central, inPress. ⟨hal-03293375⟩



Record views


Files downloads