A Clinical Benchmark of Public Self-Supervised Pathology Foundation Models

TL;DR

This paper establishes a comprehensive benchmark for evaluating public self-supervised pathology models across diverse clinical tasks, aiding in model selection and development for computational pathology applications.

Contribution

It introduces a new collection of pathology datasets and systematically assesses public foundation models, providing insights for future model training and selection.

Findings

Public models vary significantly in performance across tasks.

Certain models excel in specific organ or disease contexts.

Benchmark results guide best practices for model development.

Abstract

The use of self-supervised learning (SSL) to train pathology foundation models has increased substantially in the past few years. Notably, several models trained on large quantities of clinical data have been made publicly available in recent months. This will significantly enhance scientific research in computational pathology and help bridge the gap between research and clinical deployment. With the increase in availability of public foundation models of different sizes, trained using different algorithms on different datasets, it becomes important to establish a benchmark to compare the performance of such models on a variety of clinically relevant tasks spanning multiple organs and diseases. In this work, we present a collection of pathology datasets comprising clinical slides associated with clinically relevant endpoints including cancer diagnoses and a variety of biomarkers generated during standard hospital operation from two medical centers. We leverage these datasets to systematically assess the performance of public pathology foundation models and provide insights into best practices for training new foundation models and selecting appropriate pretrained models.