Benchmark compares data-free model merging methods for cancer survival prediction
Benchmarking Data-Free Continual Model Merging for Whole-Slide Image Survival Analysis
Computer Vision and Pattern Recognition
Summary
Predicting cancer patient survival from whole-slide images requires models that learn from new patient groups without forgetting earlier knowledge. The authors study different ways to combine models trained on separate groups without needing old data. They test six merging methods on several cancer types and find that no single method is clearly best. Their work helps others develop better ways to keep improving survival analysis models over time.
What this means in practice
- •For medical imaging teams: Combine models trained on different cancer patient groups for survival prediction without accessing old data sets.
- •For healthcare ai developers: Improve continual learning systems for cancer prognosis models deployed in hospitals by comparing model merging techniques.
Authors
Vu Minh Tran, Doanh C. Bui, Maï K. Nguyen, Khang Nguyen
Abstract
Continual survival analysis on whole-slide images (WSIs) requires models to integrate knowledge from sequentially arriving cancer cohorts while preserving previously learned information. Model merging offers a data-free alternative by combining independently fine-tuned task experts without retaining previous training data. However, the relative behavior of different merging strategies for WSI-based survival prediction remains insufficiently understood. In this work, we establish an end-to-end benchmark for data-free continual model merging, covering task-specific fine-tuning, sequential merging, and inference under both TASK-IL and CLASS-IL settings. We systematically evaluate six representative merging strategies across four TCGA cancer cohorts and three task orders using C-index, IPCW C-index, Forgetting, and Backward Transfer under a controlled protocol. Results show that no merging strategy consistently dominates across all evaluation criteria, while censoring-aware evaluation can substantially alter the relative ranking of the compared methods. This benchmark provides a unified foundation for developing and evaluating more effective continual model merging methods for WSI-based survival analysis.