Preventing dataset shift from breaking machine-learning biomarkers

作者：Jérôme Dockès, Gaël Varoquaux, Jean‐Baptiste Poline · 发表于：GigaScience · 年份：2021 · DOI：10.1093/gigascience/giab055 · 被引用次数：100 · 研究领域：Artificial Intelligence in Healthcare and Education、Artificial Intelligence in Healthcare、Machine Learning in Healthcare

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 break machine-learning-extracted biomarkers, as well as detection and correction strategies.