Point-supervised Brain Tumor Segmentation with Box-prompted MedSAM

TL;DR

This paper introduces an iterative, point-supervised segmentation framework leveraging MedSAM and semantic box prompts to improve brain tumor segmentation accuracy, reducing annotation costs and enhancing model performance.

Contribution

It presents a novel semantic-aware point supervision method with iterative refinement, enabling effective utilization of point annotations for medical image segmentation.

Findings

Outperforms traditional point-supervised methods in brain tumor segmentation.

Achieves comparable results to box-supervised methods on BraTS2018.

Demonstrates iterative framework improves segmentation accuracy progressively.

Abstract

Delineating lesions and anatomical structure is important for image-guided interventions. Point-supervised medical image segmentation (PSS) has great potential to alleviate costly expert delineation labeling. However, due to the lack of precise size and boundary guidance, the effectiveness of PSS often falls short of expectations. Although recent vision foundational models, such as the medical segment anything model (MedSAM), have made significant advancements in bounding-box-prompted segmentation, it is not straightforward to utilize point annotation, and is prone to semantic ambiguity. In this preliminary study, we introduce an iterative framework to facilitate semantic-aware point-supervised MedSAM. Specifically, the semantic box-prompt generator (SBPG) module has the capacity to convert the point input into potential pseudo bounding box suggestions, which are explicitly refined by the prototype-based semantic similarity. This is then succeeded by a prompt-guided spatial refinement (PGSR) module that harnesses the exceptional generalizability of MedSAM to infer the segmentation mask, which also updates the box proposal seed in SBPG. Performance can be progressively improved with adequate iterations. We conducted an evaluation on BraTS2018 for the segmentation of whole brain tumors and demonstrated its superior performance compared to traditional PSS methods and on par with box-supervised methods.