Class-Dependent Hybrid Data Augmentation for Multiclass Migraine Classification under Severe Class Imbalance
For researchers working on medical time-series classification under severe class imbalance, this work provides a reproducible evaluation framework and a hybrid augmentation strategy, though the main improvement is attributed to clinically motivated class aggregation rather than the augmentation method itself.
The paper corrects methodological flaws in prior migraine classification studies, achieving a corrected macro-F1 baseline of 0.71, and proposes a class-dependent hybrid augmentation framework that improves average macro-F1 across eight classifiers from 0.801 (no augmentation) to 0.862, with a peak of 0.914 using FT-Transformer.
We conducted a reproducibility-oriented re-evaluation of prior migraine classification studies, correcting for data leakage and metric bias. We then introduced (i) a clinically motivated aggregation of two hemiplegic subtypes following ICHD-3 §1.2.3, (ii) a class-dependent hybrid augmentation strategy that assigns generation methods based on per-class sample size, and (iii) the concept of fidelity asymmetry, motivating proportionally constrained growth as an alternative to full class balance. Experiments were performed on a dataset of 400 patients across seven migraine subtypes under a two-stage protocol, including the six-class configuration described above. Models were evaluated using stratified 5-fold cross-validation with macro-averaged F1 as the primary metric. Correcting methodological flaws reduces previously inflated performance estimates, with the corrected macro-F1 baseline standing at 0.71. The proposed framework consistently outperformed individual augmenters in macro-F1 averaged across the eight evaluated classifiers (0.862 vs. 0.836 for Gaussian Copula, 0.815 for CTGAN, and 0.801 for the no-augmentation baseline), and achieved its peak result of 0.914 with FT-Transformer under proportional augmentation. The no-augmentation FT-Transformer baseline (0.896) shows that, at the per-classifier ceiling, clinically motivated class aggregation accounts for most of the absolute improvement; the framework's principal measurable contribution is the gain in average robustness across classifiers, highlighting the dominant role of problem formulation.