GBT-SAM: A Parameter-Efficient Depth-Aware Model for Generalizable Brain tumour Segmentation on mp-MRI
GBT-SAM adapts Segment Anything Model (SAM) for multi-parametric MRI (mp-MRI) brain tumor segmentation using a parameter-efficient approach. It combines processing all four MRI modalities, incorporates depth information to model inter-slice relationships, and uses LoRA blocks for efficient fine-tuning. This results in state-of-the-art performance with minimal trainable parameters.