Is Synthetic Image Augmentation Useful for Imbalanced Classification Problems? Case-Study on the MIDOG2025 Atypical Cell Detection Competition
This addresses a clinically relevant but imbalanced classification problem in medical imaging, with incremental contributions to model selection and data augmentation strategies.
The study tackled the problem of classifying atypical mitotic figures in histopathology images, a highly imbalanced task, by comparing two backbones (ConvNeXt-Small and a domain-specific ViT) and synthetic data augmentation, finding that both models achieved strong performance (mean AUROC ~95%) but synthetic balancing did not consistently improve results.
The MIDOG 2025 challenge extends prior work on mitotic figure detection by introducing a new Track 2 on atypical mitosis classification. This task aims to distinguish normal from atypical mitotic figures in histopathology images, a clinically relevant but highly imbalanced and cross-domain problem. We investigated two complementary backbones: (i) ConvNeXt-Small, pretrained on ImageNet, and (ii) a histopathology-specific ViT from Lunit trained via self-supervision. To address the strong prevalence imbalance (9408 normal vs. 1741 atypical), we synthesized additional atypical examples to approximate class balance and compared models trained with real-only vs. real+synthetic data. Using five-fold cross-validation, both backbones reached strong performance (mean AUROC approximately 95 percent), with ConvNeXt achieving slightly higher peaks while Lunit exhibited greater fold-to-fold stability. Synthetic balancing, however, did not lead to consistent improvements. On the organizers' preliminary hidden test set, explicitly designed as an out-of-distribution debug subset, ConvNeXt attained the highest AUROC (95.4 percent), whereas Lunit remained competitive on balanced accuracy. These findings suggest that both ImageNet and domain-pretrained backbones are viable for atypical mitosis classification, with domain-pretraining conferring robustness and ImageNet pretraining reaching higher peaks, while naive synthetic balancing has limited benefit. Full hidden test set results will be reported upon challenge completion.