SAGE: Agentic Framework for Interpretable and Clinically Translatable Computational Pathology Biomarker Discovery
Year: 2026
Type: preprint
Abstract: Engineered image-based biomarkers offer a clinically interpretable alternative to black-box AI in computational pathology, yet their discovery remains largely intuition-driven, guided by fragmented literature rather than rigorous biological validation. We introduce SAGE (Structured Agentic system for hypothesis Generation and Evaluation), a multi-agent framework that grounds biomarker discovery in biological evidence through three mechanisms: (i) knowledge-graph-anchored hypothesis generation via multi-path ontological reasoning, (ii) a debate-based multi-agent novelty assessment that stress-tests candidate biomarkers against existing literature, and (iii) an end-to-end automated validation pipeline that translates hypotheses directly into executable analyses on multimodal pathology datasets. Together, these components shift biomarker discovery from an intuition-driven, literature-browsing exercise into a structured, traceable reasoning process that clinicians and researchers can inspect, trust, and build upon.
Source: arXiv (Cornell University)
Authors: Nasser, Sahar Almahfouz, Borja, Juan Francisco Pesantez, Liu, Jincheng, Manandhar, Sandeep, Shiromani, Shikhar, Hasan, Mohammad Tanvir, Wang, Zenghan, Ghosh, Suman, Li, Jinchu, Xu, Xuejian, Iyer, Aniket Ramkrishnan, Tokuyama, Naoto, Shah, Twisha, Pathak, Tilak, Kumaresan, Soundharya, Abe, Yohei, Maurya, Himanshu, Madabhushi, Anant
Topic: AI in cancer detection
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7127538665
Cites:
Cited by:
Related to: