Open Access

ARTICLE

CT Segmentation of Liver and Tumors Fused Multi-Scale Features

by Aihong Yu¹, Zhe Liu^1,*, Victor S. Sheng², Yuqing Song¹, Xuesheng Liu³, Chongya Ma⁴, Wenqiang Wang¹, Cong Ma¹

1 School of Computer Science and Communications Engineering, Jiangsu University, Zhenjiang, 212013, China
2 Department of Computer Science, University of Central Arkansas, Conway, 72032, USA
3 Department of Anesthesiology, the First Affiliated Hospital of Anhui Medical University, Hefei, 230001, China
4 College of Tourism and Geographic Science, Jilin Normal University, Jilin, 136000, China

* Corresponding Author: Zhe Liu. Email:

Intelligent Automation & Soft Computing 2021, 30(2), 589-599. https://doi.org/10.32604/iasc.2021.019513

Received 14 April 2021; Accepted 15 May 2021; Issue published 11 August 2021

Abstract

Liver cancer is one of frequent causes of death from malignancy in the world. Owing to the outstanding advantages of computer-aided diagnosis and deep learning, fully automatic segmentation of computed tomography (CT) images turned into a research hotspot over the years. The liver has quite low contrast with the surrounding tissues, together with its lesion areas are thoroughly complex. To deal with these problems, we proposed effective methods for enhancing features and processed public datasets from Liver Tumor Segmentation Challenge (LITS) for the verification. In this experiment, data pre-processing based on the image enhancement and noise reduction. This study redesigned the original UNet with two novel modules and named it DResUNet which was applied deformable convolution. The first module aimed to recalibrate information by the channel and spatial dimension. The other module enriched deep information of these liver CT images through fusing multi-scale features. Besides, we used cross-entropy loss function for adaptive weights to solve the troubles of class imbalance in the dataset samples. These can improve the performance of the network in-depth and breadth feature learning to deal with many complex segmentation scenes in abdominal CT images. More importantly, the effect of predicted images fully proved that our methods are highly competitive among the segmentation of liver and liver tumors.

---

Keywords

---