Probing Intrinsic Medical Task Relationships: A Contrastive Learning Perspective

TL;DR

This paper investigates the intrinsic relationships between 30 diverse medical vision tasks across multiple datasets using a novel contrastive learning framework called TaCo, revealing how tasks relate and differ in a shared representation space.

Contribution

The introduction of Task-Contrastive Learning (TaCo), a new contrastive learning framework that embeds heterogeneous medical vision tasks into a unified representation space.

Findings

TaCo effectively maps tasks from different modalities into a joint space.

The analysis reveals which tasks are distinctly represented and which are similar.

Iterative task alterations are reflected in the embedding space.

Abstract

While much of the medical computer vision community has focused on advancing performance for specific tasks, the underlying relationships between tasks, i.e., how they relate, overlap, or differ on a representational level, remain largely unexplored. Our work explores these intrinsic relationships between medical vision tasks, specifically, we investigate 30 tasks, such as semantic tasks (e.g., segmentation and detection), image generative tasks (e.g., denoising, inpainting, or colorization), and image transformation tasks (e.g., geometric transformations). Our goal is to probe whether a data-driven representation space can capture an underlying structure of tasks across a variety of 39 datasets from wildly different medical imaging modalities, including computed tomography, magnetic resonance, electron microscopy, X-ray ultrasound and more. By revealing how tasks relate to one another, we aim to provide insights into their fundamental properties and interconnectedness. To this end, we introduce Task-Contrastive Learning (TaCo), a contrastive learning framework designed to embed tasks into a shared representation space. Through TaCo, we map these heterogeneous tasks from different modalities into a joint space and analyze their properties: identifying which tasks are distinctly represented, which blend together, and how iterative alterations to tasks are reflected in the embedding space. Our work provides a foundation for understanding the intrinsic structure of medical vision tasks, offering a deeper understanding of task similarities and their interconnected properties in embedding spaces.