MPath: Multimodal Pathology Report Generation from Whole Slide Images

TL;DR

MPath is a novel multimodal framework that generates pathology reports from whole slide images by conditioning a pretrained language model with visual embeddings, demonstrating promising results on a challenging dataset.

Contribution

Introduces MPath, a lightweight multimodal approach using visual-prefix prompting to generate pathology reports from WSIs without end-to-end training.

Findings

Ranked 4th in RED 2025 Challenge

Effective use of foundation-model WSI features

Prompt-based conditioning improves report generation

Abstract

Automated generation of diagnostic pathology reports directly from whole slide images (WSIs) is an emerging direction in computational pathology. Translating high-resolution tissue patterns into clinically coherent text remains difficult due to large morphological variability and the complex structure of pathology narratives. We introduce MPath, a lightweight multimodal framework that conditions a pretrained biomedical language model (BioBART) on WSI-derived visual embeddings through a learned visual-prefix prompting mechanism. Instead of end-to-end vision-language pretraining, MPath leverages foundation-model WSI features (CONCH + Titan) and injects them into BioBART via a compact projection module, keeping the language backbone frozen for stability and data efficiency. MPath was developed and evaluated on the RED 2025 Grand Challenge dataset and ranked 4th in Test Phase 2, despite limited submission opportunities. The results highlight the potential of prompt-based multimodal conditioning as a scalable and interpretable strategy for pathology report generation.