Synthetic Data Generation for Long-Tail Medical Image Classification: A Case Study in Skin Lesions
For medical image classification, this work addresses the critical problem of rare disease diagnosis by improving performance on tail classes, though it is an incremental application of diffusion models to a known bottleneck.
The paper tackles long-tail medical image classification by introducing a diffusion-model-driven synthetic data augmentation pipeline with an inpainting diffusion model and OOD post-selection. On the ISIC2019 skin lesion dataset, it achieves over 28% improvement on the tail class with the fewest samples.
Long-tailed class distributions are pervasive in multi-class medical datasets and pose significant challenges for deep learning models which typically underperform on tail classes with limited samples. This limitation is particularly problematic in medical applications, where rare classes often correspond to severe or high-risk diseases and therefore require high diagnostic accuracy. Existing solutions-including specialized architectures, rebalanced loss functions, and handcrafted data augmentation-offer only marginal improvements and struggle to scale due to their limited and largely deterministic variability. To address these challenges, we introduce a diffusion-model-driven synthetic data augmentation pipeline tailored for medical long-tailed classification. Our approach features a novel inpainting diffusion model combined with an Out-of-Distribution (OOD) post-selection mechanism to ensure diverse, realistic, and clinically meaningful synthetic samples. Evaluated on the ISIC2019 skin lesion classification dataset, one of the largest and most imbalanced medical imaging benchmarks, our method yields substantial improvements in overall performance, with particularly pronounced gains on tail classes with more than $28\%$ improvement on the class with the fewest samples. These results demonstrate the effectiveness of diffusion-based augmentation in mitigating long-tail imbalance and enhancing medical classification robustness.