Evislide mil improves cancer subtype recognition in whole-slide images
EviSlide-MIL: Multi-Query Pooling and Shared Prediction for Class-Incremental Whole-Slide Image Classification
Computer Vision and Pattern Recognition
Summary
Classifying cancer types from large microscope images is tricky because new types keep appearing, and models often forget old types when learning new ones. The authors propose EviSlide-MIL, a system that uses special queries to remember features of known cancer classes while learning new ones without forgetting. Their method freezes parts of the model related to old classes and updates shared parts carefully, helping it keep good performance across many cancer types. Tests on tumor data showed it works better than earlier methods at remembering and recognizing many cancer subtypes over time.
What this means in practice
- •For clinical histopathology teams: Enable pathology labs to update cancer subtype classifiers incrementally without losing accuracy on previous diagnoses.
- •For medical imaging software developers: Build scalable WSI analysis tools that handle new tumor classes flexibly while preserving existing model knowledge.
Authors
Zhouhan Qian, Sicheng Chen, Chad Wong, Tianyi Zhang, Enhui Chai, Zeyu Liu, Fei Xia
Abstract
Histopathological analysis of whole-slide images (WSIs) plays a critical role in cancer diagnosis, and multiple instance learning (MIL) is widely adopted for WSI analysis under weak supervision. As new tumor subtypes are introduced, extending existing MIL models requires learning new classes while retaining knowledge of previously learned ones. This presents a class-incremental learning challenge, where models must acquire new classes with limited access to historical training data and recognize all learned classes without task identifiers at inference. A major obstacle is forgetting, as updates to feature aggregation and prediction modules can degrade performance on previously learned classes. To address this challenge, we propose \textbf{EviSlide-MIL}, a class-incremental WSI framework that integrates multiple learnable queries per class with a shared prediction module. The queries aggregate frozen patch features into class-specific representations, while the shared prediction module limits parameter growth across incremental stages. By freezing queries associated with previously learned classes, EviSlide-MIL preserves their attention patterns for identical inputs, while experience replay and knowledge distillation further mitigate forgetting as shared parameters are updated. Experiments on eight tumor subtypes from The Cancer Genome Atlas demonstrate that EviSlide-MIL achieves improved performance and reduced forgetting compared with competing methods across most learning settings, demonstrating its potential for scalable and knowledge-preserving WSI classification. Code and pretrained models will be released.