Efficient Knowledge Distillation of SAM for Medical Image Segmentation
This work addresses the problem of deploying efficient segmentation models in real-time, resource-constrained medical environments, representing an incremental improvement.
The paper tackled the high computational requirements of the Segment Anything Model (SAM) for medical image segmentation by proposing a knowledge distillation method, KD SAM, which achieved comparable or superior performance with significantly fewer parameters on datasets like Kvasir-SEG and ISIC 2017.
The Segment Anything Model (SAM) has set a new standard in interactive image segmentation, offering robust performance across various tasks. However, its significant computational requirements limit its deployment in real-time or resource-constrained environments. To address these challenges, we propose a novel knowledge distillation approach, KD SAM, which incorporates both encoder and decoder optimization through a combination of Mean Squared Error (MSE) and Perceptual Loss. This dual-loss framework captures structural and semantic features, enabling the student model to maintain high segmentation accuracy while reducing computational complexity. Based on the model evaluation on datasets, including Kvasir-SEG, ISIC 2017, Fetal Head Ultrasound, and Breast Ultrasound, we demonstrate that KD SAM achieves comparable or superior performance to the baseline models, with significantly fewer parameters. KD SAM effectively balances segmentation accuracy and computational efficiency, making it well-suited for real-time medical image segmentation applications in resource-constrained environments.