A Clinical-grade Universal Foundation Model for Intraoperative Pathology

2025-10-06

AI Summary

Intraoperative pathological diagnosis is hindered by suboptimal frozen-section quality, limited annotated data, and poor generalizability—impeding clinical deployment of AI. To address this, we introduce CRISP, the first clinical-grade foundation model for intraoperative pathology, trained on 100,000 multi-center, multi-cancer-type, and multi-anatomic-site frozen-section images. CRISP achieves robust zero-shot generalization to unseen anatomic sites and rare cancer types—previously unattained. Validated on retrospective and prospective dual cohorts (n > 2,000), it supports real-time benign-malignant classification, critical intraoperative decision-making, and pan-cancer detection: directly guiding surgical decisions in 92.6% of cases, reducing pathologist workload by 35%, eliminating 105 unnecessary ancillary tests, and achieving 87.5% accuracy in micrometastasis detection. This work advances AI-driven routine intraoperative pathology.

Technology Category

• Computer Vision: Other Foundations of Computer Vision
• Machine Learning: Other Foundations of Machine Learning
• Humans and AI: Other Foundations of Human Computation & AI

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Responsible Web: Algorithmic accountability and transparency on the web

Abstract

Intraoperative pathology is pivotal to precision surgery, yet its clinical impact is constrained by diagnostic complexity and the limited availability of high-quality frozen-section data. While computational pathology has made significant strides, the lack of large-scale, prospective validation has impeded its routine adoption in surgical workflows. Here, we introduce CRISP, a clinical-grade foundation model developed on over 100,000 frozen sections from eight medical centers, specifically designed to provide Clinical-grade Robust Intraoperative Support for Pathology (CRISP). CRISP was comprehensively evaluated on more than 15,000 intraoperative slides across nearly 100 retrospective diagnostic tasks, including benign-malignant discrimination, key intraoperative decision-making, and pan-cancer detection, etc. The model demonstrated robust generalization across diverse institutions, tumor types, and anatomical sites-including previously unseen sites and rare cancers. In a prospective cohort of over 2,000 patients, CRISP sustained high diagnostic accuracy under real-world conditions, directly informing surgical decisions in 92.6% of cases. Human-AI collaboration further reduced diagnostic workload by 35%, avoided 105 ancillary tests and enhanced detection of micrometastases with 87.5% accuracy. Together, these findings position CRISP as a clinical-grade paradigm for AI-driven intraoperative pathology, bridging computational advances with surgical precision and accelerating the translation of artificial intelligence into routine clinical practice.

Problem

• Developing a universal AI model for intraoperative pathology diagnosis
• Addressing diagnostic complexity and limited frozen-section data availability
• Validating clinical-grade AI performance across diverse surgical scenarios

Innovation

• Clinical-grade foundation model using 100,000 frozen sections
• Validated on 15,000 slides across 100 diagnostic tasks
• Human-AI collaboration reduces workload by 35%