Simplex-Aligned Diffusion with Cross-Granularity Interaction for Robust Medical Image Classification

Chao Wu, Mingchen Gao
Proceedings of The 9th International Conference on Medical Imaging with Deep Learning, PMLR 315:121-152, 2026.

Abstract

The clinical deployment of medical image classification systems hinges on their trustworthiness, specifically, the ability to provide calibrated uncertainty estimates and maintain robustness under acquisition shifts. While generative diffusion models offer promising distributional modeling, existing approaches suffer from a fundamental geometric conflict: they apply unbounded Gaussian noise directly to bounded label simplices. We identify that this theoretical mismatch forces predictions into invalid probability spaces, serving as a primary source of model unreliability and overconfidence. To resolve this, we propose Simplex-Aligned Diffusion. Unlike standard methods, we reformulate the label generation process on an unconstrained logit manifold. By mapping the probability simplex to a Euclidean space, we ensure mathematical consistency with Gaussian diffusion, which effectively acts as a geometric regularizer for uncertainty calibration. Furthermore, we introduce a Transformer-based Cross-Granularity Interaction module to stabilize visual guidance by dynamically modeling global-local dependencies. Extensive experiments on the APTOS2019 and HAM10000 benchmarks demonstrate that our framework not only achieves competitive accuracy but significantly outperforms state-of-the-art baselines in calibration error (ECE) and resilience to clinical artifacts (e.g., sensor noise, blur), offering a mathematically rigorous and clinically reliable paradigm.Code is available at

Cite this Paper

BibTeX
@InProceedings{pmlr-v315-wu26a, title = {Simplex-Aligned Diffusion with Cross-Granularity Interaction for Robust Medical Image Classification}, author = {Wu, Chao and Gao, Mingchen}, booktitle = {Proceedings of The 9th International Conference on Medical Imaging with Deep Learning}, pages = {121--152}, year = {2026}, editor = {Huo, Yuankai and Gao, Mingchen and Kuo, Chang-Fu and Jin, Yueming and Deng, Ruining}, volume = {315}, series = {Proceedings of Machine Learning Research}, month = {08--10 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v315/main/assets/wu26a/wu26a.pdf}, url = {https://proceedings.mlr.press/v315/wu26a.html}, abstract = {The clinical deployment of medical image classification systems hinges on their trustworthiness, specifically, the ability to provide calibrated uncertainty estimates and maintain robustness under acquisition shifts. While generative diffusion models offer promising distributional modeling, existing approaches suffer from a fundamental geometric conflict: they apply unbounded Gaussian noise directly to bounded label simplices. We identify that this theoretical mismatch forces predictions into invalid probability spaces, serving as a primary source of model unreliability and overconfidence. To resolve this, we propose Simplex-Aligned Diffusion. Unlike standard methods, we reformulate the label generation process on an unconstrained logit manifold. By mapping the probability simplex to a Euclidean space, we ensure mathematical consistency with Gaussian diffusion, which effectively acts as a geometric regularizer for uncertainty calibration. Furthermore, we introduce a Transformer-based Cross-Granularity Interaction module to stabilize visual guidance by dynamically modeling global-local dependencies. Extensive experiments on the APTOS2019 and HAM10000 benchmarks demonstrate that our framework not only achieves competitive accuracy but significantly outperforms state-of-the-art baselines in calibration error (ECE) and resilience to clinical artifacts (e.g., sensor noise, blur), offering a mathematically rigorous and clinically reliable paradigm.Code is available at } }
Endnote
%0 Conference Paper %T Simplex-Aligned Diffusion with Cross-Granularity Interaction for Robust Medical Image Classification %A Chao Wu %A Mingchen Gao %B Proceedings of The 9th International Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2026 %E Yuankai Huo %E Mingchen Gao %E Chang-Fu Kuo %E Yueming Jin %E Ruining Deng %F pmlr-v315-wu26a %I PMLR %P 121--152 %U https://proceedings.mlr.press/v315/wu26a.html %V 315 %X The clinical deployment of medical image classification systems hinges on their trustworthiness, specifically, the ability to provide calibrated uncertainty estimates and maintain robustness under acquisition shifts. While generative diffusion models offer promising distributional modeling, existing approaches suffer from a fundamental geometric conflict: they apply unbounded Gaussian noise directly to bounded label simplices. We identify that this theoretical mismatch forces predictions into invalid probability spaces, serving as a primary source of model unreliability and overconfidence. To resolve this, we propose Simplex-Aligned Diffusion. Unlike standard methods, we reformulate the label generation process on an unconstrained logit manifold. By mapping the probability simplex to a Euclidean space, we ensure mathematical consistency with Gaussian diffusion, which effectively acts as a geometric regularizer for uncertainty calibration. Furthermore, we introduce a Transformer-based Cross-Granularity Interaction module to stabilize visual guidance by dynamically modeling global-local dependencies. Extensive experiments on the APTOS2019 and HAM10000 benchmarks demonstrate that our framework not only achieves competitive accuracy but significantly outperforms state-of-the-art baselines in calibration error (ECE) and resilience to clinical artifacts (e.g., sensor noise, blur), offering a mathematically rigorous and clinically reliable paradigm.Code is available at
APA
Wu, C. & Gao, M.. (2026). Simplex-Aligned Diffusion with Cross-Granularity Interaction for Robust Medical Image Classification. Proceedings of The 9th International Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 315:121-152 Available from https://proceedings.mlr.press/v315/wu26a.html.

Related Material