Agent-Guided Relational Concept Discovery: Toward Interpretable Surgical Margin Assessment

2026-07-23

AI Summary

This study addresses the challenges of poor generalization and limited interpretability in deep learning models for intraoperative margin assessment, particularly under noisy conditions and with unlabeled data. The authors propose a proxy-guided concept discovery framework that operates without predefined concept labels, integrating a reasoning agent mechanism with biochemical knowledge graph embeddings in an unsupervised setting for the first time. The reasoning agent refines semantic concept descriptions, while the knowledge graph imposes biologically grounded relational constraints. Evaluated on REIMS datasets for skin and breast cancer, the method significantly improves balanced accuracy and sensitivity, yields fewer false positives in intraoperative settings, demonstrates superior generalization, and enhances model interpretability.

Abstract

Deep learning models can effectively use Rapid Evaporative Ionization Mass Spectrometry (REIMS) data for surgical margin assessment. However, their clinical adoption remains challenging due to limited generalization to operating room conditions. This difficulty arises because models are typically trained on labeled spectra collected from resected tissue samples, while they must operate on noisy, unlabeled data acquired directly during surgery. In addition, the black-box nature of deep learning models makes it difficult to understand and systematically improve their behavior. Concept-based learning offers a promising way to address these challenges by mapping raw measurements to human-understandable concepts. However, supervised concept-based approaches rely on concept annotations, which are difficult to obtain in complex mass spectrometry workflows. We propose Agent-Guided Concept Discovery, a framework that learns meaningful concepts directly from data without requiring predefined concept labels. During training, a reasoning agent refines semantic descriptions of the learned concepts and adaptively adjusts their weight based on diagnostic relevance. These concepts are further grounded using a biochemical knowledge graph to ensure consistency with known metabolic relationships. Across Skin and Breast Cancer datasets, our model improves balanced accuracy and sensitivity over the baseline. In a representative intraoperative case, it shows fewer false positives, indicating better generalization to surgical conditions.

Problem

• surgical margin assessment
• concept-based learning
• REIMS
• generalization
• interpretability

Innovation

• Agent-Guided Concept Discovery
• Interpretable AI
• REIMS
• Surgical Margin Assessment
• Knowledge Graph