LINGUAL: Language-INtegrated GUidance in Active Learning for Medical Image Segmentation

TL;DR

LINGUAL introduces a language-guided framework for medical image segmentation active learning, significantly reducing annotation effort by translating natural language instructions into executable tasks, and achieving comparable or better results than traditional methods.

Contribution

It presents a novel language-integrated guidance system for active learning in medical image segmentation, enabling minimal expert effort and automatic task execution.

Findings

Achieves comparable or superior performance to active learning baselines.

Reduces annotation time by approximately 80%.

Demonstrates effectiveness in active domain adaptation.

Abstract

Although active learning (AL) in segmentation tasks enables experts to annotate selected regions of interest (ROIs) instead of entire images, it remains highly challenging, labor-intensive, and cognitively demanding due to the blurry and ambiguous boundaries commonly observed in medical images. Also, in conventional AL, annotation effort is a function of the ROI- larger regions make the task cognitively easier but incur higher annotation costs, whereas smaller regions demand finer precision and more attention from the expert. In this context, language guidance provides an effective alternative, requiring minimal expert effort while bypassing the cognitively demanding task of precise boundary delineation in segmentation. Towards this goal, we introduce LINGUAL: a framework that receives natural language instructions from an expert, translates them into executable programs through in-context learning, and automatically performs the corresponding sequence of sub-tasks without any human intervention. We demonstrate the effectiveness of LINGUAL in active domain adaptation (ADA) achieving comparable or superior performance to AL baselines while reducing estimated annotation time by approximately 80%.