Agent Aggregator with Mask Denoise Mechanism for Histopathology Whole Slide Image Analysis
This work improves medical diagnosis for pathologists by enhancing accuracy in cancer detection from histopathology images, though it is incremental as it builds on existing MIL approaches.
The paper tackles the challenge of classifying whole slide images in histopathology by addressing limitations of attention mechanisms in multiple instance learning, proposing AMD-MIL which achieves superior performance over state-of-the-art methods on datasets like CAMELYON-16 and TCGA-LUNG.
Histopathology analysis is the gold standard for medical diagnosis. Accurate classification of whole slide images (WSIs) and region-of-interests (ROIs) localization can assist pathologists in diagnosis. The gigapixel resolution of WSI and the absence of fine-grained annotations make direct classification and analysis challenging. In weakly supervised learning, multiple instance learning (MIL) presents a promising approach for WSI classification. The prevailing strategy is to use attention mechanisms to measure instance importance for classification. However, attention mechanisms fail to capture inter-instance information, and self-attention causes quadratic computational complexity. To address these challenges, we propose AMD-MIL, an agent aggregator with a mask denoise mechanism. The agent token acts as an intermediate variable between the query and key for computing instance importance. Mask and denoising matrices, mapped from agents-aggregated value, dynamically mask low-contribution representations and eliminate noise. AMD-MIL achieves better attention allocation by adjusting feature representations, capturing micro-metastases in cancer, and improving interpretability. Extensive experiments on CAMELYON-16, CAMELYON-17, TCGA-KIDNEY, and TCGA-LUNG show AMD-MIL's superiority over state-of-the-art methods.