Think as Cardiac Sonographers: Marrying SAM with Left Ventricular Indicators Measurements According to Clinical Guidelines

TL;DR

This paper introduces AutoSAME, a novel framework that combines the Segment Anything Model with specialized techniques to accurately measure left ventricular indicators in echocardiography, mimicking expert cardiac sonographers.

Contribution

AutoSAME integrates SAM with landmark localization and segmentation, introducing FCBA and SGPA to improve accuracy in LV measurements according to clinical guidelines.

Findings

AutoSAME outperforms existing methods in LV segmentation.

It achieves high accuracy in landmark localization.

The framework demonstrates superior LV indicator measurement performance.

Abstract

Left ventricular (LV) indicator measurements following clinical echocardiog-raphy guidelines are important for diagnosing cardiovascular disease. Alt-hough existing algorithms have explored automated LV quantification, they can struggle to capture generic visual representations due to the normally small training datasets. Therefore, it is necessary to introduce vision founda-tional models (VFM) with abundant knowledge. However, VFMs represented by the segment anything model (SAM) are usually suitable for segmentation but incapable of identifying key anatomical points, which are critical in LV indicator measurements. In this paper, we propose a novel framework named AutoSAME, combining the powerful visual understanding of SAM with seg-mentation and landmark localization tasks simultaneously. Consequently, the framework mimics the operation of cardiac sonographers, achieving LV indi-cator measurements consistent with clinical guidelines. We further present fil-tered cross-branch attention (FCBA) in AutoSAME, which leverages relatively comprehensive features in the segmentation to enhance the heatmap regression (HR) of key points from the frequency domain perspective, optimizing the vis-ual representation learned by the latter. Moreover, we propose spatial-guided prompt alignment (SGPA) to automatically generate prompt embeddings guid-ed by spatial properties of LV, thereby improving the accuracy of dense pre-dictions by prior spatial knowledge. The extensive experiments on an echocar-diography dataset demonstrate the efficiency of each design and the superiori-ty of our AutoSAME in LV segmentation, landmark localization, and indicator measurements. The code will be available at https://github.com/QC-LIU-1997/AutoSAME.