Xiang-Bin Luo

SAKD-SAM: efficient breast tumor MRI segmentation via self-attention knowledge distillation and 3D sparse flash attention

Segmenting breast tumors in magnetic resonance imaging (MRI) is a critical step in computer-aided diagnosis. Recently, the Segment Anything Model (SAM) has demonstrated remarkable potential in medical image segmentation. However, the immense parameter size and heavy computational burden of SAM's Transformer-based archi...