MergeSurv: Merging-Based Continual Learning for Survival Analysis on Whole-Slide Images
For computational pathology, this work addresses the need for scalable and privacy-preserving continual learning across cancer cohorts, though the novelty is incremental as it applies model merging to a specific domain.
MergeSurv proposes a merging-based continual learning framework for survival analysis on whole-slide images, enabling adaptation to multiple cancer cohorts without storing previous data. It outperforms naive fine-tuning and other continual learning methods on four TCGA cohorts, reducing catastrophic forgetting.
Survival analysis on Whole Slide Images (WSIs) is important in computational pathology for prognosis estimation and treatment planning. However, existing survival models are typically trained independently for each cancer cohort, making continual adaptation computationally expensive for gigapixel-scale WSIs. In this study, we propose MergeSurv, a merging-based continual learning framework for WSI survival analysis. A pathology vision-language foundation model is independently fine-tuned on each task, and the learned parameters are sequentially merged into a unified model without storing previous training data. We further investigate two inference strategies: One-for-All (OFA) and Voting-Expert Aggregation (VEA). Experiments on four TCGA cohorts demonstrate that MergeSurv outperforms naive fine-tuning as well as representative regularization-based and rehearsal-based continual learning methods, while effectively reducing catastrophic forgetting. The results suggest that model merging is a promising direction for scalable and privacy-preserving continual learning in computational pathology.