TY - JOUR
T1 - Model drift
T2 - When it can be a sign of success and when it can be an occult problem
AU - Carter, Rickey E.
AU - Anand, Vidhu
AU - Harmon, David M.
AU - Pellikka, Patricia A.
N1 - Publisher Copyright:
© 2022 The Authors
PY - 2022/1
Y1 - 2022/1
N2 - The practice of medicine is currently prioritizing the development, validation and implementation of machine learning algorithms. Measuring how well the algorithms function upon implementation in the practice is an area of critical need. When changes in model performance occur after implementation, a concept generally labeled model drift, careful examination of the putative reasons for the change is needed. This article presents two scenarios that illustrate how model drift can indicate favorable and unfavorable changes in performance.
AB - The practice of medicine is currently prioritizing the development, validation and implementation of machine learning algorithms. Measuring how well the algorithms function upon implementation in the practice is an area of critical need. When changes in model performance occur after implementation, a concept generally labeled model drift, careful examination of the putative reasons for the change is needed. This article presents two scenarios that illustrate how model drift can indicate favorable and unfavorable changes in performance.
UR - http://www.scopus.com/inward/record.url?scp=85127467037&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85127467037&partnerID=8YFLogxK
U2 - 10.1016/j.ibmed.2022.100058
DO - 10.1016/j.ibmed.2022.100058
M3 - Article
AN - SCOPUS:85127467037
VL - 6
JO - Intelligence-Based Medicine
JF - Intelligence-Based Medicine
SN - 2666-5212
M1 - 100058
ER -