A Latent Adversarial Cauchy-Schwarz Autoencoder for Medical Image Segmentation

Medical image segmentation plays a vital role in clinical diagnosis. However, previous methods cannot handle intrinsic ambiguities in extracting deep semantics of medical images. Moreover, they neglect fruitful semantic information in segmentation maps. To address the challenges, a latent adversarial Cauchy-Schwarz autoencoder is proposed, which defines image segmentation as a cross-modal translation task from medical images to segmentation maps. Specifically, a probabilistic graph model is defined to fit the conditional distribution of the image translation between modalities, which leverages the Cauchy-Schwarz divergence to alleviate approximation errors caused by ambiguities in extracting deep semantics. Then, a novel numerical solution is derived to optimize the probabilistic graph model, which explores semantics in segmentation maps to facilitate the segmentation. Afterwards, a dual-flow architecture is proposed with an adversarial encoding-decoding paradigm to implement the numerical solution. Finally, extensive experiments in two medical scenarios illustrate that the proposed method achieves the state-of-the-art performance compared with nine baseline methods.

I. Introduction

Medical image segmentation, as a fundamental technique in clinical diagnoses, leverages pixel-wise information to split images into multiple non-overlapping regions where pixels in the same region are more similar than pixels in different regions [1], [2]. It segments crucial lesions and abnormal tissues in medical images to provide accurate anatomical structures and pathologies in organs of interest for physicians, which can greatly shorten many time-consuming clinical diagnosis processes, such as disease evaluation and treatment planning.