SegmentWithSAM: 3D Slicer Extension for Segment Anything Model (SAM)

Zafer Yildiz, Hanxue Gu, Jikai Zhang, Jichen Yang, Maciej A Mazurowski

Keywords: Segmentation, SAM, 3D Slicer

Abstract: The development of reliable automated deep learning-based algorithms for the segmentation of medical images is heavily reliant on training data in the form of images along with outlines of the objects of interest. However, manual annotation of medical images is very time-consuming and often requires highly specialized expertise. Here, we provide software that incorporates the recently developed and highly impactful Segment Anything Model (SAM) into the popular software for the visualization and annotation of medical images, 3D Slicer. SAM has been developed to segment any object with prompt-based user guidance. It has been shown to be successful in aiding some annotations in medical imaging. The software described in this paper allows to leverage the power of SAM while using the highly convenient and publicly available 3D Slicer software. Our code is publicly available on https://github.com/mazurowski-lab/SlicerSegmentWithSAM, and it can be installed directly from the Extension Manager of 3D Slicer.

Submission Number: 149