A Reproducible Framework for Bias-Resistant Machine Learning on Small-Sample Neuroimaging Data
This provides a generalizable blueprint for reliable machine learning in data-limited biomedical domains, though it is incremental as it builds on existing methods like nested cross-validation.
The paper tackled the problem of optimistic bias in cross-validation for small-sample neuroimaging data by introducing a reproducible framework, achieving a nested-CV balanced accuracy of 0.660 ± 0.068 on a structural MRI dataset.
We introduce a reproducible, bias-resistant machine learning framework that integrates domain-informed feature engineering, nested cross-validation, and calibrated decision-threshold optimization for small-sample neuroimaging data. Conventional cross-validation frameworks that reuse the same folds for both model selection and performance estimation yield optimistically biased results, limiting reproducibility and generalization. Demonstrated on a high-dimensional structural MRI dataset of deep brain stimulation cognitive outcomes, the framework achieved a nested-CV balanced accuracy of 0.660\,$\pm$\,0.068 using a compact, interpretable subset selected via importance-guided ranking. By combining interpretability and unbiased evaluation, this work provides a generalizable computational blueprint for reliable machine learning in data-limited biomedical domains.