LIMIS: Towards Language-based Interactive Medical Image Segmentation

TL;DR

LIMIS is a novel language-based interactive segmentation model for medical images that enables radiologists to refine segmentation masks solely through natural language commands, enhancing usability and efficiency.

Contribution

This work introduces the first purely language-based interactive segmentation model for medical images, adapting grounded models and designing a new interaction strategy.

Findings

High-quality segmentation masks produced

Effective user interaction through language commands

Validated on three medical datasets with expert confirmation

Abstract

Within this work, we introduce LIMIS: The first purely language-based interactive medical image segmentation model. We achieve this by adapting Grounded SAM to the medical domain and designing a language-based model interaction strategy that allows radiologists to incorporate their knowledge into the segmentation process. LIMIS produces high-quality initial segmentation masks by leveraging medical foundation models and allows users to adapt segmentation masks using only language, opening up interactive segmentation to scenarios where physicians require using their hands for other tasks. We evaluate LIMIS on three publicly available medical datasets in terms of performance and usability with experts from the medical domain confirming its high-quality segmentation masks and its interactive usability.