Machine learning can easily produce false positives when the test set is wrongly used. Just et al in @NatureHumBehav suggested that ML can identify suicidal ideation extremely well from fMRI and we were skeptical. Today retraction and our analysis of what went wrong came out.
Teacher, Clinical Supervisor, [AIdvisor]
Retired Clinical Associate Professor, Psychiatry, University of Florida
Affiliation is for identification only. All content and opinions are my own.
View more posts
Gator Psychiatry 