Automatic liver tumour segmentation in CT combining FCN and NMF-based deformable model

ABSTRACT

Automatic liver tumour segmentation is an important step towards digital medical research, clinical diagnosis and therapy planning. However, the existence of noise, low contrast and heterogeneity make the automatic liver tumour segmentation remaining an open challenge. In this work, we focus on a novel automatic method to segment liver tumour in abdomen images from CT scans using fully convolutional networks (FCN) and non-negative matrix factorization (NMF) based deformable model. We train the FCN for semantic liver and tumour segmentation using preprocessed training data by BM3D. The segmented liver and tumour regions of FCN are used as ROI and initialization for the NMF-based tumour refinement, respectively. We refine the surfaces of tumours using a 3D deformable model which derived from NMF and driven by local cumulative spectral histograms (LCSH). The refinement is designed to obtain a smoother, more accurate and natural liver tumour surface. Experiments on 11 clinical datasets demonstrated that the proposed segmentation method achieves satisfactory results.

KEYWORDS:

• Liver tumor
• FCN
• non-negative matrix factorization
• local cumulative spectral histograms
• BM3D

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. https://competitions.codalab.org/competitions/17094.

2. http://www.ircad.fr/research/3d-ircadb-01/.

3. https://github.com/shekkizh/FCN.tensorflow.

Additional information

Funding

This research is sponsored by the National Natural Science Foundation of China [grant number 61876026, 61472053 and 61672120].

Notes on contributors

Shenhai Zheng

Shenhai Zheng received the B.E. degree in Mathematics and Applied Mathematics from Hubei Minzu University, Enshi, China, the M.S. and Ph.D degree in Computational Mathematics and Computer Science and Technology from Chongqing University. He is currently a teacher in the College of Computer Science and Technology, Chongqing University of Posts and Telecommunications. His research interests include intelligent computation, pattern recognition, image processing and machine vision.

Bin Fang

Bin Fang received the B.E. degree in Electrical Engineering from Xian Jiaotong University, Xian, China, the M.S. degree in Electrical Engineering from Sichuan University, Chengdu, China, and the Ph.D. degree in Electrical Engineering from the University of Hong Kong, Hong Kong, China. He is currently a professor in the College of Computer Science, Chongqing University. His research interests include computer vision, pattern recognition, medical image processing, biometrics applications, and document analysis.

Laquan Li

Laquan Li is a teacher in the College of Science, Chongqing University of Posts and Telecommunications, China. She received her PhD degree from the School of Automation, Huazhong University of Science and Technology, China. Her research interests include medical image processing and analysis, and variational method for inverse problem.

Mingqi Gao

Mingqi Gao received his M.S. degree from the School of Computer Science, Chongqing University, China and B.S. degree from Inner Mongolia University, China in 2014.

Yi Wang

Yi Wang received his M.S. degree and Ph.D degree from the College of Mechanical Engineering and Computer Science and Technology from Chongqing University. Chongqing University, China. He is currently working as a telaboratory technician at the College of Computer Science, Chongqing University. His research interests include pattern recognition, medical image processing and machine vision.

Kaiyi Peng

Kaiyi Peng obtained his Master degree major in Computer Science and Technology in Sichuan Normal University, China. Currently, he is a Ph.D. candidate in College of Computer Science, Chongqing University. His research interests lie in intelligent computation, pattern recognition, image processing and machine vision.