Quilt-1M: One Million Image-Text Pairs for Histopathology

TL;DR

This paper introduces QUILT-1M, the largest vision-language dataset for histopathology with one million image-text pairs, enabling improved representation learning and classification in medical imaging.

Contribution

The creation of QUILT-1M, a large-scale histopathology image-text dataset, and demonstrating its effectiveness in improving model performance on classification and retrieval tasks.

Findings

Outperforms state-of-the-art models on classification tasks

Enables effective zero-shot learning for histopathology images

Improves cross-modal retrieval accuracy

Abstract

Recent accelerations in multi-modal applications have been made possible with the plethora of image and text data available online. However, the scarcity of analogous data in the medical field, specifically in histopathology, has slowed comparable progress. To enable similar representation learning for histopathology, we turn to YouTube, an untapped resource of videos, offering $1,087$ hours of valuable educational histopathology videos from expert clinicians. From YouTube, we curate QUILT: a large-scale vision-language dataset consisting of $802, 144$ image and text pairs. QUILT was automatically curated using a mixture of models, including large language models, handcrafted algorithms, human knowledge databases, and automatic speech recognition. In comparison, the most comprehensive datasets curated for histopathology amass only around $200$K samples. We combine QUILT with datasets from other sources, including Twitter, research papers, and the internet in general, to create an even larger dataset: QUILT-1M, with $1$M paired image-text samples, marking it as the largest vision-language histopathology dataset to date. We demonstrate the value of QUILT-1M by fine-tuning a pre-trained CLIP model. Our model outperforms state-of-the-art models on both zero-shot and linear probing tasks for classifying new histopathology images across $13$ diverse patch-level datasets of $8$ different sub-pathologies and cross-modal retrieval tasks.