Paired Uterine Whole-Slide Images and Pathology Reports for Multimodal Computational Pathology

2026-07-04

AI Summary

This study addresses a critical limitation in uterine pathology research—the scarcity of paired whole-slide image (WSI) and clinical report datasets—by introducing TUM-Uteria, the first multimodal uterine pathology dataset offering dual-granularity alignment at both case and slide levels. Comprising 216 cases from a tertiary medical center and 455 high-quality WSI–report pairs, the dataset was curated through a structured acquisition pipeline and rigorously validated by multiple pathologists across several stages to ensure clinical accuracy and annotation consistency. TUM-Uteria establishes a foundational resource for advancing AI-driven pathological research, particularly in whole-slide image analysis, multimodal learning, and automated diagnostic report generation.

Abstract

Uterine diseases represent an important category of gynecologic pathology and require accurate histopathological assessment for diagnosis and treatment planning. Whole-slide images (WSI) have enabled the digital transformation of pathology workflows and provided new opportunities for artificial intelligence (AI) in computational pathology. In particular, multimodal models that jointly analyze histopathology images and pathology reports have shown promising potential for automated pathology report generation and AI-assisted diagnosis. However, the development of such systems remains limited by the scarcity of datasets that pair whole-slide images with clinically meaningful pathology reports. Instead, existing pathology datasets focus on patch- or slide-level annotations of a single endpoint (e.g., disease class), which do not fully capture the rich information in full clinical diagnostic workflow reports. Here, we introduce TUM-Uteria, a uterine pathology dataset comprising WSIs paired with diagnostic pathology reports at both the case and slide levels, collected from a tertiary medical center. The dataset contains 216 clinical cases, comprising 455 slide-level WSI-report pairs. The dataset underwent a structured multi-stage validation procedure involving board-certified pathologists to ensure reliable annotations. TUM-Uteria supports research in computational pathology, including whole-slide image analysis, multimodal learning, and automated pathology report generation.

Problem

• whole-slide images
• pathology reports
• multimodal computational pathology
• uterine diseases
• dataset scarcity

Innovation

• whole-slide image
• multimodal learning
• pathology report generation
• computational pathology
• uterine pathology