Continual Multiple Instance Learning with Enhanced Localization for Histopathological Whole Slide Image Analysis

Byung Hyun Lee, Wongi Jeong, Woojae Han, KYOUNGBUN LEE, Se Young Chun

Abstract: Multiple instance learning (MIL) significantly reduced annotation costs via bag-level weak labels for large-scale images, such as histopathological whole slide images (WSIs); however, its adaptability to continual tasks with minimal forgetting has been rarely explored, especially on instance classification for localization. Weakly incremental learning for semantic segmentation has been studied for continual localization, but it focused on natural images, leveraging global relationships among hundreds of small patches using pre-trained models—an approach that seems infeasible for MIL localization due to enormous amounts of large patches and a lack of available global relationships. To address these challenges, we propose Continual Multiple Instance Learning with Enhanced Localization (CoMEL), an MIL framework consisting of a Grouped Double Attention Transformer (GDAT) for efficient encoding, Bag Prototypes-based Pseudo-Labeling (BPPL) for reliable labeling, and Orthogonal Weighted Low-Rank Adaptation (OWLoRA) to mitigate forgetting, with extensive experiments on three public WSI datasets demonstrating that CoMEL outperforms prior arts by up to 11.00% in bag-level accuracy and 23.4% in localization accuracy.