The newly released Segment Anything Model (SAM) is a popular tool used in image processing due to its superior segmentation accuracy, input prompts, training capabilities, and efficient model design. However, its current model isn’t tailored to medical images, specifically ultrasound images, which tend to have a lot of noise. This project developed ClickSAM, which fine-tunes SAM using click prompts for ultrasound images. ClickSAM has two stages of training. The first stage is trained on single-click prompts centered in the ground-truth contours, and the second stage focuses on improving the model performance through additional positive and negative click prompts. ClickSAM exhibits superior performance compared to other existing models for ultrasound image segmentation.
Publication date: 9 Feb 2024
Project Page: Not Given
Paper: https://arxiv.org/pdf/2402.05902
