MorphSAM: Learning the Morphological Prompts from Atlases for Spine Image Segmentation

TL;DR

This paper introduces MorphSAM, a novel approach that enhances spine image segmentation by explicitly learning morphological prompts from atlases and integrating them into the Segment Anything Model, improving accuracy in clinical applications.

Contribution

The paper presents a new method that explicitly learns morphological information from atlases and incorporates it into SAM for improved spine segmentation performance.

Findings

MorphSAM outperforms state-of-the-art methods in spine segmentation tasks.

It effectively captures morphological features from atlases.

Experimental validation on CT and MR images confirms its superiority.

Abstract

Spine image segmentation is crucial for clinical diagnosis and treatment of spine diseases. The complex structure of the spine and the high morphological similarity between individual vertebrae and adjacent intervertebral discs make accurate spine segmentation a challenging task. Although the Segment Anything Model (SAM) has been proposed, it still struggles to effectively capture and utilize morphological information, limiting its ability to enhance spine image segmentation performance. To address these challenges, in this paper, we propose a MorphSAM that explicitly learns morphological information from atlases, thereby strengthening the spine image segmentation performance of SAM. Specifically, the MorphSAM includes two fully automatic prompt learning networks, 1) an anatomical prompt learning network that directly learns morphological information from anatomical atlases, and 2) a semantic prompt learning network that derives morphological information from text descriptions converted from the atlases. Then, the two learned morphological prompts are fed into the SAM model to boost the segmentation performance. We validate our MorphSAM on two spine image segmentation tasks, including a spine anatomical structure segmentation task with CT images and a lumbosacral plexus segmentation task with MR images. Experimental results demonstrate that our MorphSAM achieves superior segmentation performance when compared to the state-of-the-art methods.