Interactive 3D Medical Image Segmentation with SAM 2

TL;DR

This paper investigates the use of SAM 2's zero-shot capabilities for 3D medical image segmentation by treating slices as video frames, enabling automatic annotation propagation and reducing manual effort.

Contribution

It introduces a practical pipeline leveraging SAM 2 for 3D medical segmentation, demonstrating its effectiveness and potential to narrow the gap with supervised methods.

Findings

SAM 2 can propagate annotations across 3D volumes effectively.

The method reduces annotation effort significantly.

SAM 2 shows promising results on BraTS2020 and Decathlon datasets.

Abstract

Interactive medical image segmentation (IMIS) has shown significant potential in enhancing segmentation accuracy by integrating iterative feedback from medical professionals. However, the limited availability of enough 3D medical data restricts the generalization and robustness of most IMIS methods. The Segment Anything Model (SAM), though effective for 2D images, requires expensive semi-auto slice-by-slice annotations for 3D medical images. In this paper, we explore the zero-shot capabilities of SAM 2, the next-generation Meta SAM model trained on videos, for 3D medical image segmentation. By treating sequential 2D slices of 3D images as video frames, SAM 2 can fully automatically propagate annotations from a single frame to the entire 3D volume. We propose a practical pipeline for using SAM 2 in 3D medical image segmentation and present key findings highlighting its efficiency and potential for further optimization. Concretely, numerical experiments on the BraTS2020 and the medical segmentation decathlon datasets demonstrate that SAM 2 still has a gap with supervised methods but can narrow the gap in specific settings and organ types, significantly reducing the annotation burden on medical professionals. Our code will be open-sourced and available at https://github.com/Chuyun-Shen/SAM_2_Medical_3D.