Agentic Visual Reasoning in Whole-Slide Pathology Images via Active Perception

2026-08-09

AI Summary

This work addresses the challenge of efficiently localizing sparse diagnostic evidence and integrating morphological associations across scales in gigapixel whole-slide images (WSIs). To this end, the authors propose AdaptivePath, a framework that formulates evidence acquisition as a sequential decision-making process. It employs a Navigator module for active perception from low to high magnification, coupled with a Morphology Interpreter and a Deliberator to perform question-guided evidence extraction and iterative answer refinement. The approach introduces an anomaly-driven navigation strategy that requires no task-specific trajectory annotations, alternately optimizing representation learning and perception policy. A geometry-appearance consistency objective is further designed to stabilize focus trajectories, enabling traceable visual reasoning. AdaptivePath achieves zero-shot state-of-the-art performance on WSI visual question answering and six-class TCGA cancer subtype classification, attaining 80.14% accuracy and improving pathologists’ diagnostic accuracy to 82.9%.

Abstract

Whole-slide visual reasoning requires identifying sparse diagnostic evidence in gigapixel pathology slides and integrating observations across spatial scales. Existing WSI methods either compress densely sampled patches into global representations or use pretrained vision-language models with heuristic region selection, weakening links between predictions and morphology or lacking pathology-trained observation policies. We present AdaptivePath, an active-perception framework that formulates WSI evidence acquisition as sequential decision making. The Navigator learns question-agnostic abnormality-driven navigation from pathologist-reviewed labels to select observation locations and spatial extents, avoiding costly question-specific trajectory annotations. We train this policy through alternating representation learning and proximal policy optimization, followed by fine-tuning with geometric and appearance consistency objectives to stabilize focus trajectories. During inference, the Navigator hierarchically acquires sparse observations from low to high magnification under a limited ROI budget. A Morphology Interpreter converts observations into question-conditioned evidence, while the Deliberator evaluates evidence and revises intermediate answers across magnifications. The Arbiter integrates deliberation history to produce final answers. AdaptivePath achieves state-of-the-art zero-shot performance on WSI and region pathology VQA benchmarks and reaches 80.14% accuracy for cancer subtype classification across six TCGA cohorts. In a blinded diagnostic-utility study, pathologists using AdaptivePath-selected observation sequences achieve 82.9% accuracy. These results demonstrate that learned active perception enables effective and traceable visual reasoning over gigapixel pathology slides.

Problem

• whole-slide imaging
• visual reasoning
• active perception
• pathology diagnosis
• gigapixel images

Innovation

• active perception
• whole-slide imaging
• visual reasoning
• reinforcement learning
• pathology AI