Assessing Foundational Medical 'Segment Anything' (Med-SAM1, Med-SAM2) Deep Learning Models for Left Atrial Segmentation in 3D LGE MRI

TL;DR

This study evaluates the effectiveness of foundational deep learning models, MedSAM1 and MedSAM2, in automating left atrium segmentation in 3D LGE MRI for atrial fibrillation assessment, comparing their accuracy and prompting strategies.

Contribution

It provides the first evaluation of MedSAM models for complex 3D cardiac MRI segmentation, highlighting differences between single-prompt and multi-prompt approaches.

Findings

MedSAM2 with automated tracking outperforms MedSAM1 in segmentation accuracy.

Performance varies with prompt size and location, affecting Dice scores.

MedSAM models show promise for efficient, zero-shot cardiac MRI segmentation.

Abstract

Atrial fibrillation (AF), the most common cardiac arrhythmia, is associated with heart failure and stroke. Accurate segmentation of the left atrium (LA) in 3D late gadolinium-enhanced (LGE) MRI is helpful for evaluating AF, as fibrotic remodeling in the LA myocardium contributes to arrhythmia and serves as a key determinant of therapeutic strategies. However, manual LA segmentation is labor-intensive and challenging. Recent foundational deep learning models, such as the Segment Anything Model (SAM), pre-trained on diverse datasets, have demonstrated promise in generic segmentation tasks. MedSAM, a fine-tuned version of SAM for medical applications, enables efficient, zero-shot segmentation without domain-specific training. Despite the potential of MedSAM model, it has not yet been evaluated for the complex task of LA segmentation in 3D LGE-MRI. This study aims to (1) evaluate the performance of MedSAM in automating LA segmentation, (2) compare the performance of the MedSAM2 model, which uses a single prompt with automated tracking, with the MedSAM1 model, which requires separate prompt for each slice, and (3) analyze the performance of MedSAM1 in terms of Dice score(i.e., segmentation accuracy) by varying the size and location of the box prompt.