Generalize Ultrasound Image Segmentation via Instant and Plug & Play Style Transfer
This work provides a fast and lightweight solution for robust ultrasound image segmentation under unknown appearance shifts, which is crucial for practical clinical adoption.
The paper addresses the challenge of generalizing deep segmentation models to ultrasound images with unknown appearance shifts, which typically require retraining and lead to high latency. The authors propose a one-stage plug-and-play solution that embeds hierarchical style transfer units and uses Dynamic Instance Normalization to simultaneously remove appearance shifts and perform segmentation. This method adds only 0.2ms and 1.92M FLOPs for a 400x400 image, demonstrating enhanced robustness across a large dataset from three vendors.
Deep segmentation models that generalize to images with unknown appearance are important for real-world medical image analysis. Retraining models leads to high latency and complex pipelines, which are impractical in clinical settings. The situation becomes more severe for ultrasound image analysis because of their large appearance shifts. In this paper, we propose a novel method for robust segmentation under unknown appearance shifts. Our contribution is three-fold. First, we advance a one-stage plug-and-play solution by embedding hierarchical style transfer units into a segmentation architecture. Our solution can remove appearance shifts and perform segmentation simultaneously. Second, we adopt Dynamic Instance Normalization to conduct precise and dynamic style transfer in a learnable manner, rather than previously fixed style normalization. Third, our solution is fast and lightweight for routine clinical adoption. Given 400*400 image input, our solution only needs an additional 0.2ms and 1.92M FLOPs to handle appearance shifts compared to the baseline pipeline. Extensive experiments are conducted on a large dataset from three vendors demonstrate our proposed method enhances the robustness of deep segmentation models.