Generalize Ultrasound Image Segmentation Via Instant And Plug & Play Style Transfer

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 0.2 ms 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.

1. Introduction

The tremendous success of deep neural networks (DNNs) has benefitted medical image analysis [1]. However, deployment of DNN models in real clinical scenarios is threatened by appearance shifts that degrade their performance. Different sources of appearance variation affect routine medical image acquisition, including operators, protocols, vendors, parameters and tissue properties, all of which can lead to unpredictable image appearance changes [2, 3]. The adverse effect of appearance shift on ultrasound image segmentation can be observed in Fig.1. Making DNNs robust against appearance shift is along the last mile before they can be clinically adopted.