Consortium-driven development of pathology foundation model-based approaches for automated scoring of histopathology in ulcerative colitis

Study Background

We utilized a consortia approach to address the lack of reproducibility for histological assessment in ulcerative colitis (UC) clinical trials by joining
stakeholders and contributors from academia, biopharma, government, non-profits, and industry.

With the goal of improving histological endpoint assessment in clinical trials, the Foundation for National Institutes of Health Biomarkers

Consortium Mucosal Healing in UC Project Team established a cross-industry framework of pooled real-world and clinical trial patient data.

Here, we report the results of this consortium-wide effort to standardize and automate Geboes score (GS), leveraging machine-learning (ML)-based
whole slide image (WSI) predictions of GS grades, subgrades, and key thresholds for histologic improvement and remission.

Conference

Digestive Disease Week 2025

Partner

FINH