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.
@article{arxiv.2602.00953,
title = {SAGE: Agentic Framework for Interpretable and Clinically Translatable Computational Pathology Biomarker Discovery},
author = {Sahar Almahfouz Nasser and Juan Francisco Pesantez Borja and Jincheng Liu and Sandeep Manandhar and Shikhar Shiromani and Mohammad Tanvir Hasan and Zenghan Wang and Suman Ghosh and Jinchu Li and Xuejian Xu and Aniket Ramkrishnan Iyer and Naoto Tokuyama and Twisha Shah and Tilak Pathak and Soundharya Kumaresan and Yohei Abe and Himanshu Maurya and Anant Madabhushi},
journal= {arXiv preprint arXiv:2602.00953},
year = {2026}
}