Abstract

Info

Title:: Liver segmentation method based on multi-scale feature fusion and attention

Author(s):: RAN Meizi1; HU Xiaojun2; JIANG Xiaoyan1; FAN Yingfang2; WANG Hang1; WANG Hailing1; GAO Yongbin1; 1. School of Electronic and Electrical Engineering, Shanghai University of Engineering Science, Shanghai 201620, China 2. Department of Hepatobiliary Surgery, the Fifth Affiliated Hospital of Southern Medical University, Guangzhou 510000, China

Keywords:: Keywords: liver segmentation attention mechanism multi-scale feature fusion deep supervision MFFA UNet

Abstract:: Abstract: Due to the low contrast of CT images, irregular shape of the liver, and blurred boundaries with adjacent organs, the existing methods based on convolutional neural network underperform in liver segmentation tasks, especially for boundary recognition and small object detection. A novel liver segmentation method is proposed based on multi-scale feature fusion and attention, namely MFFA UNet. Multi-scale feature fusion is firstly employed to acquire abundant segmentation details, while spatial and channel attention mechanisms are utilized to capture global spatial and inter-channel relationships. Additionally, a deep supervision module fully leverages the output of intermediate hidden layers, enhancing the learning capability of the network, which in turn accelerates the networks convergence speed. Moreover, a hybrid loss function is adopted to address the issue of class imbalance, further boosting the models segmentation efficacy. Experimental results demonstrate that the proposed MFFA UNet outperforms the prevailing segmentation networks on the public LITS dataset, producing results that are closer to the ground truth.