MultiTalent: A Multi-Dataset Approach to Medical Image Segmentation

TL;DR

MultiTalent introduces a multi-dataset training approach for medical image segmentation, leveraging diverse annotated CT datasets to improve performance and serve as a powerful foundation model.

Contribution

It presents a novel method to train a single model on multiple datasets with conflicting labels, enhancing segmentation accuracy and pre-training effectiveness.

Findings

Improved segmentation performance over single dataset models.

Effective handling of conflicting class definitions across datasets.

Superior pre-training benefits for various segmentation tasks.

Abstract

The medical imaging community generates a wealth of datasets, many of which are openly accessible and annotated for specific diseases and tasks such as multi-organ or lesion segmentation. Current practices continue to limit model training and supervised pre-training to one or a few similar datasets, neglecting the synergistic potential of other available annotated data. We propose MultiTalent, a method that leverages multiple CT datasets with diverse and conflicting class definitions to train a single model for a comprehensive structure segmentation. Our results demonstrate improved segmentation performance compared to previous related approaches, systematically, also compared to single dataset training using state-of-the-art methods, especially for lesion segmentation and other challenging structures. We show that MultiTalent also represents a powerful foundation model that offers a superior pre-training for various segmentation tasks compared to commonly used supervised or unsupervised pre-training baselines. Our findings offer a new direction for the medical imaging community to effectively utilize the wealth of available data for improved segmentation performance. The code and model weights will be published here: [tba]