Priority-Aware Pathological Hierarchy Training for Multiple Instance Learning

2025-07-27

AI Summary

Existing clinical multiple instance learning (MIL) methods neglect the clinically grounded priority relationships among pathological manifestations and diagnostic categories, leading to misclassification of critical severe conditions. Method: We propose a novel two-level (vertical + horizontal) MIL training framework that explicitly models the hierarchical priority structure among diagnostic categories for the first time. It incorporates cross-level prediction alignment and implicit feature reuse to enhance model sensitivity to severe pathologies without additional annotation cost. The method integrates multi-instance modeling, a hierarchical loss function, and cross-layer consistency optimization. Results: Evaluated on real-world patient data, our approach significantly reduces misdiagnosis rates and markedly improves robustness and clinical applicability in detecting key pathological features—particularly in complex, multi-symptom cases.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Computer Vision: Multi-modal Vision
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Multiple Instance Learning (MIL) is increasingly being used as a support tool within clinical settings for pathological diagnosis decisions, achieving high performance and removing the annotation burden. However, existing approaches for clinical MIL tasks have not adequately addressed the priority issues that exist in relation to pathological symptoms and diagnostic classes, causing MIL models to ignore priority among classes. To overcome this clinical limitation of MIL, we propose a new method that addresses priority issues using two hierarchies: vertical inter-hierarchy and horizontal intra-hierarchy. The proposed method aligns MIL predictions across each hierarchical level and employs an implicit feature re-usability during training to facilitate clinically more serious classes within the same level. Experiments with real-world patient data show that the proposed method effectively reduces misdiagnosis and prioritizes more important symptoms in multiclass scenarios. Further analysis verifies the efficacy of the proposed components and qualitatively confirms the MIL predictions against challenging cases with multiple symptoms.

Problem

• Addresses priority issues in pathological symptoms and diagnostic classes
• Aligns MIL predictions across hierarchical levels for clinical accuracy
• Reduces misdiagnosis by prioritizing important symptoms in multiclass scenarios

Innovation

• Priority-aware hierarchical training for MIL
• Vertical and horizontal hierarchy alignment
• Implicit feature re-usability for serious classes