Towards SAMBA: Segment Anything Model for Brain Tumor Segmentation in Sub-Saharan African Populations

Gliomas, the most prevalent primary brain tumors, require precise segmentation for diagnosis and treatment planning. However, this task poses significant challenges, particularly in the African population, where limited access to high-quality imaging data hampers algorithm performance. In this study, we propose a new approach combining the Segment Anything Model (SAM) and a voting network for multi-modal glioma segmentation. By fine-tuning SAM with bounding box-guided prompts (SAMBA), we adapt the model to the complexities of African datasets. Our ensemble strategy, utilizing multiple modalities and views, produces a robust consensus segmentation, addressing the intratumoral heterogeneity. This study was conducted on the Brain Tumor Segmentation (BraTS) Africa (BraTS-Africa) dataset, which provides a valuable resource for addressing challenges specific to resource-limited settings and facilitating the development of effective and more generalizable segmentation algorithms. To illustrate our approach’s potential, our experiments on the BraTS-Africa dataset yielded compelling results, with SAMBA attaining a Dice coefficient of 86.6% for binary segmentation and 60.4% for multi-class segmentation. Although the low quality of the scans currently presents difficulties, SAMBA has the potential to facilitate more generalizable segmentations for real world clinical problems with future applications to other types of brain lesions.