Multimodal Whole Slide Foundation Model for Pathology

TL;DR

This paper introduces TITAN, a multimodal foundation model for pathology that leverages extensive self-supervised and vision-language training on whole slide images and reports, enabling effective clinical task performance without fine-tuning.

Contribution

The paper presents TITAN, a novel multimodal foundation model trained on large-scale pathology data, capable of generalizing to resource-limited clinical scenarios without fine-tuning.

Findings

TITAN outperforms existing models in classification and retrieval tasks.

TITAN effectively generates pathology reports without clinical labels.

TITAN demonstrates strong zero-shot and few-shot learning capabilities.

Abstract

The field of computational pathology has been transformed with recent advances in foundation models that encode histopathology region-of-interests (ROIs) into versatile and transferable feature representations via self-supervised learning (SSL). However, translating these advancements to address complex clinical challenges at the patient and slide level remains constrained by limited clinical data in disease-specific cohorts, especially for rare clinical conditions. We propose TITAN, a multimodal whole slide foundation model pretrained using 335,645 WSIs via visual self-supervised learning and vision-language alignment with corresponding pathology reports and 423,122 synthetic captions generated from a multimodal generative AI copilot for pathology. Without any finetuning or requiring clinical labels, TITAN can extract general-purpose slide representations and generate pathology reports that generalize to resource-limited clinical scenarios such as rare disease retrieval and cancer prognosis. We evaluate TITAN on diverse clinical tasks and find that TITAN outperforms both ROI and slide foundation models across machine learning settings such as linear probing, few-shot and zero-shot classification, rare cancer retrieval and cross-modal retrieval, and pathology report generation.