Optimization of MedSAM model based on bounding box adaptive perturbation algorithm
This work addresses segmentation challenges in medical imaging, particularly for small or irregular targets, but appears incremental as it builds on the existing MedSAM framework.
The study tackled limitations in the MedSAM model for medical image segmentation, such as errors in small tissue segmentation and poor performance with reduced bounding box prompts, by proposing a bounding box adaptive perturbation algorithm to improve robustness and reliability.
The MedSAM model, built upon the SAM framework, enhances medical image segmentation through generalizable training but still exhibits notable limitations. First, constraints in the perturbation window settings during training can cause MedSAM to incorrectly segment small tissues or organs together with adjacent structures, leading to segmentation errors. Second, when dealing with medical image targets characterized by irregular shapes and complex structures, segmentation often relies on narrowing the bounding box to refine segmentation intent. However, MedSAM's performance under reduced bounding box prompts remains suboptimal. To address these challenges, this study proposes a bounding box adaptive perturbation algorithm to optimize the training process. The proposed approach aims to reduce segmentation errors for small targets and enhance the model's accuracy when processing reduced bounding box prompts, ultimately improving the robustness and reliability of the MedSAM model for complex medical imaging tasks.