Gaze-Assisted Medical Image Segmentation

TL;DR

This paper introduces a semi-supervised medical image segmentation method that uses human gaze data to improve segmentation accuracy, fine-tuning MedSAM for better clinical applicability.

Contribution

It demonstrates that incorporating human gaze as a prompt in MedSAM significantly enhances segmentation performance on abdominal CT scans.

Findings

Gaze-assisted MedSAM achieved an average Dice coefficient of 90.5%.

Outperformed state-of-the-art segmentation models like nnUNetV2 and ResUNet.

Validated on a public dataset of 120 CT scans with 16 organs.

Abstract

The annotation of patient organs is a crucial part of various diagnostic and treatment procedures, such as radiotherapy planning. Manual annotation is extremely time-consuming, while its automation using modern image analysis techniques has not yet reached levels sufficient for clinical adoption. This paper investigates the idea of semi-supervised medical image segmentation using human gaze as interactive input for segmentation correction. In particular, we fine-tuned the Segment Anything Model in Medical Images (MedSAM), a public solution that uses various prompt types as additional input for semi-automated segmentation correction. We used human gaze data from reading abdominal images as a prompt for fine-tuning MedSAM. The model was validated on a public WORD database, which consists of 120 CT scans of 16 abdominal organs. The results of the gaze-assisted MedSAM were shown to be superior to the results of the state-of-the-art segmentation models. In particular, the average Dice coefficient for 16 abdominal organs was 85.8%, 86.7%, 81.7%, and 90.5% for nnUNetV2, ResUNet, original MedSAM, and our gaze-assisted MedSAM model, respectively.