Self-Reinforcing For Few-Shot Medical Image Segmentation

Abstract

Few-shot learning serves as a viable solution for addressing data scarcity, thus exhibiting significant potential in the domain of medical image segmentation. In this work, we propose a simple and efficient framework for few-shot medical image segmentation, termed SRPNet, which leverages self-reinforcement between foreground and background. Notably, without the need for prior knowledge, the model autonomously adapts the segmentation effect of both foreground and background, thereby enhancing the segmentation of previously unseen classes. Experimental evaluations conducted on CT and MRI datasets demonstrate the superior performance of the proposed method compared to other state-of-the-art techniques. Code is available at https://github.com/q362096112/SRPNet.