This title appears in the Scientific Report :
2023
Please use the identifier:
http://dx.doi.org/10.1093/gigascience/giad071 in citations.
Please use the identifier: http://dx.doi.org/10.34734/FZJ-2023-03119 in citations.
Confound-leakage: Confound Removal in Machine Learning Leads to Leakage
Confound-leakage: Confound Removal in Machine Learning Leads to Leakage
BackgroundMachine learning (ML) approaches are a crucial component of modern data analysis in many fields, including epidemiology and medicine. Nonlinear ML methods often achieve accurate predictions, for instance, in personalized medicine, as they are capable of modeling complex relationships betwe...
Saved in:
Personal Name(s): | Hamdan, Sami |
---|---|
Love, Bradley C / Polier, Georg von / Weis, Susanne / Schwender, Holger / Eickhoff, Simon / Patil, Kaustubh (Corresponding author) | |
Contributing Institute: |
Gehirn & Verhalten; INM-7 |
Published in: | GigaScience, 12 (20323) S. giad071 |
Imprint: |
Oxford
Oxford University Press
2023
|
DOI: |
10.1093/gigascience/giad071 |
DOI: |
10.34734/FZJ-2023-03119 |
Document Type: |
Journal Article |
Research Program: |
Einzelfallvorhersagen der motorischen Fähigkeiten bei Gesunden und Patienten mit motorischen Störungen (B05) Brain Dysfunction and Plasticity |
Link: |
Get full text OpenAccess |
Publikationsportal JuSER |
Please use the identifier: http://dx.doi.org/10.34734/FZJ-2023-03119 in citations.
BackgroundMachine learning (ML) approaches are a crucial component of modern data analysis in many fields, including epidemiology and medicine. Nonlinear ML methods often achieve accurate predictions, for instance, in personalized medicine, as they are capable of modeling complex relationships between features and the target. Problematically, ML models and their predictions can be biased by confounding information present in the features. To remove this spurious signal, researchers often employ featurewise linear confound regression (CR). While this is considered a standard approach for dealing with confounding, possible pitfalls of using CR in ML pipelines are not fully understood.ResultsWe provide new evidence that, contrary to general expectations, linear confound regression can increase the risk of confounding when combined with nonlinear ML approaches. Using a simple framework that uses the target as a confound, we show that information leaked via CR can increase null or moderate effects to near-perfect prediction. By shuffling the features, we provide evidence that this increase is indeed due to confound-leakage and not due to revealing of information. We then demonstrate the danger of confound-leakage in a real-world clinical application where the accuracy of predicting attention-deficit/hyperactivity disorder is overestimated using speech-derived features when using depression as a confound.ConclusionsMishandling or even amplifying confounding effects when building ML models due to confound-leakage, as shown, can lead to untrustworthy, biased, and unfair predictions. Our expose of the confound-leakage pitfall and provided guidelines for dealing with it can help create more robust and trustworthy ML models. |