Self-supervised large-scale kidney abnormality detection in drug safety assessment studies

2025-08-29

AI Summary

In drug safety assessment, identifying rare nephrotoxic histopathological alterations from large-scale whole-slide images (WSIs) remains challenging due to the high cost and low throughput of manual review. To address this, we propose the first self-supervised learning framework for large-scale nephrotoxicity anomaly detection, leveraging the UNI foundation model to extract tissue representations and implementing a pixel-level annotation-free k-nearest neighbors (k-NN) anomaly detection scheme. Evaluated on a WSI dataset comprising kidney sections from 158 compounds, our method achieves an AUC of 0.62 and a negative predictive value of 89%, enabling efficient filtering of normal slides. Our key contribution lies in overcoming the performance bottleneck of direct foundation model classification by enabling weakly supervised, high-throughput anomaly localization tailored for preclinical toxicological screening—thereby significantly improving both efficiency and scalability of pathological evaluation.

Technology Category

• Computer Vision: Other Foundations of Computer Vision
• Machine Learning: Unsupervised & Self-Supervised Learning
• Data Mining & Knowledge Management: Anomaly/Outlier Detection

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
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Kidney abnormality detection is required for all preclinical drug development. It involves a time-consuming and costly examination of hundreds to thousands of whole-slide images per drug safety study, most of which are normal, to detect any subtle changes indicating toxic effects. In this study, we present the first large-scale self-supervised abnormality detection model for kidney toxicologic pathology, spanning drug safety assessment studies from 158 compounds. We explore the complexity of kidney abnormality detection on this scale using features extracted from the UNI foundation model (FM) and show that a simple k-nearest neighbor classifier on these features performs at chance, demonstrating that the FM-generated features alone are insufficient for detecting abnormalities. We then demonstrate that a self-supervised method applied to the same features can achieve better-than-chance performance, with an area under the receiver operating characteristic curve of 0.62 and a negative predictive value of 89%. With further development, such a model can be used to rule out normal slides in drug safety assessment studies, reducing the costs and time associated with drug development.

Problem

• Detecting kidney abnormalities in preclinical drug safety studies
• Reducing time and cost of examining numerous normal slides
• Improving self-supervised detection of subtle toxic effects

Innovation

• Self-supervised method for kidney abnormality detection
• Using UNI foundation model extracted features
• k-nearest neighbor classifier on FM features