Triplanar Convolutional Neural Network for Automatic Liver and Tumor Image Segmentation

doi:10.23940/ijpe.18.12.p24.31513158

Abstract

Abstract:

The automatic image segmentation of liver and liver tumors is important in the diagnosis and treatment of hepatocellular carcinoma. A novel triplanar fully convolutional neural network (FCN) composed of three 2D convolutional neural networks (CNNs) is proposed to handle the issue. It performs segmentation through the transverse plane, coronal plane, and sagittal plane and can effectively use multi-dimensional features for 3D segmentation. Then, a cascaded structure is used to balance the positive and negative samples for segmentation of the tumor. The experimental results are obtained through data analysis and tested on the 3DIRCADb. They show that our method outperforms the existing methods and achieves a volume overlap error of 6.7%and 3.6% on the liver and tumors respectively.

Key words: liver, liver tumors, automatic image segmentation, triplanar FCN, data analysis

Zhenggang Wang, and Guanling Wang. Triplanar Convolutional Neural Network for Automatic Liver and Tumor Image Segmentation [J]. Int J Performability Eng, 2018, 14(12): 3151-3158.

Add to citation manager EndNote|Reference Manager|ProCite|BibTeX|RefWorks

Figures/Tables 10

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Table 1

Table 2

Figure 6

Table 3

Table 4

References 21

[1]	X. Li, H. Chen, X. Qi, Q. Dou, C. W. Fu, P. A. Heng , “H-DenseUNet: Hybrid Densely Connected UNet for Liver and Tumor Segmentation from CT Volumes,” IEEE Transactions on Medical Imaging, Vol. 37, No. 12, pp. 2663-2674, 2018 doi: 10.1109/TMI.2018.2845918 pmid: 29994201
[2]	K. Simonyan and A. Zisserman, “Very Deep Convolutional Networks for Large-Scale Image Recognition, ” arXiv Preprint, arXiv: 1409. 1556, 2014
[3]	D. Morley , “Visionary Ophthalmics: Confluence of Computer Vision and Deep Learning for Ophthalmology, ” 2018
[4]	J. Schreier , “Anatomical Segmentation of CT images for Radiation Therapy planning Using Deep Learning, ” 2018
[5]	A. Sevastopolsky , “Optic Disc and Cup Segmentation Methods for Glaucoma Detection with Modification of U-Net Convolutional Neural Network,” Pattern Recognition and Image Analysis,Vol. 27, No. 3, pp. 618-624, 2017 doi: 10.1134/S1054661817030269
[6]	V. Dumoulin and F. Visin , “A Guide to Convolution Arithmetic for Deep Learning, ” arXiv Preprint, arXiv: 1603. 07285, 2016
[7]	F. C. Ghesu, B. Georgescu, S. Grbic, A. Maier, J. Hornegger, D. Comaniciu , “Towards Intelligent Robust Detection of Anatomical Structures in Incomplete Volumetric Data,” Medical Image Analysis, Vol. 48, pp. 203-213, 2018 doi: 10.1016/j.media.2018.06.007
[8]	S. Iqbal, M. U. Ghani, T. Saba, A. Rehman , “Brain Tumor Segmentation in Multi-Spectral MRI Using Convolutional Neural Networks (CNN),” Microscopy Research and Technique,Vol. 81, No. 4, pp. 419-427, 2018 doi: 10.1002/jemt.22994 pmid: 29356229
[9]	M. Ghafoorian , “Machine Learning for Quantification of Small Vessel Disease Imaging Biomarkers, ” 2018
[10]	J. Redmon and A. Farhadi,“YOLO9000: better, faster, stronger, ” arXiv Preprint, 2017 doi: 10.1109/CVPR.2017.690
[11]	J. W. Johnson, “Adapting Mask-RCNN for Automatic Nucleus Segmentation, ” arXiv Preprint, arXiv: 1805. 00500, 2018
[12]	P. Jansson , “Single-Word Speech Recognition with Convolutional Neural Networks on Raw Waveforms, ” 2018
[13]	F. Yu and V. Koltun, “Multi-Scale Context aggregation by Dilated Convolutions, ” arXiv Preprint, arXiv: 1511. 07122, 2015
[14]	P. L. Combettes and J. C. Pesquet, “Deep Neural Network Structures Solving Variational Inequalities, ” arXivPreprint arXiv: 1808. 07526, 2018
[15]	Z. Li and N. Snavely, “Learning Intrinsic Image Decomposition from Watching the World, ” arXiv Preprint, arXiv: 1804. 00582, 2018
[16]	C. Schenck and D. Fox, “Perceiving and Reasoning about Liquids Using Fully Convolutional Networks,” The International Journal of Robotics Research, Vol. 37, No.4-5, pp. 452-471, 2018 doi: 10.1177/0278364917734052
[17]	I. Goodfellow , “NIPS 2016 Tutorial: Generative Adversarial Networks,” arXiv Preprint,arXiv:1701. 00160, 2016
[18]	C. Li, X. Wang, W. Liu, L. J. Latecki , “DeepMitosis: Mitosis Detection via Deep Detection, Verification and Segmentation Networks,” Medical image Analysis, Vol. 45, pp. 121-133, 2018 doi: 10.1016/j.media.2017.12.002 pmid: 29455111
[19]	F. Xing and L. Yang ,“RobustNucleus/Cell Detection and Segmentation in Digital Pathology and Microscopy Images: A Comprehensive Review,” IEEE Reviews in Biomedical Engineering,Vol. 9, pp. 234-263, 2016 doi: 10.1109/RBME.2016.2515127 pmid: 26742143
[20]	S. Li and G. K. F. Tso. Bottleneck , “Supervised U-Net for Pixel-Wise Liver and Tumor Segmentation, ” arXiv Preprint, arXiv: 1810. 10331, 2018
[21]	Y. Wang, Y. Zhou, W. Shen, S. Park, E. K. Fishman, A. L. Yuille, “Abdominal Multi-Organ Segmentation with Organ-Attention Networks And Statistical Fusion, ” arXivPreprint, arXiv: 1804. 08414, 2018

Approach	VOE[%]	RVD[%]	ASD[mm]	MSSD[mm]
FCN YZ	10.8	4.5	4.2	16.7
FCN XZ	11.5	5.7	3.4	20.4
FCN XY	8.9	-2.3	1.2	9.3
Triplanar FCN	6.7	3.5	0.7	6.5

Approach	VOE[%]	RVD[%]	ASD[mm]	MSSD[mm]
FCN YZ	22.8	-8.3	12.3	15.7
FCN XZ	25.7	13.8	11.4	19.9
FCN XY	16.3	-3.3	2.8	8.9
Triplanar FCN	13.3	6.3	1.8	7.5

Approach	VOE [%]	RVD [%]	ASD [mm]	MSSD [mm]
Schreier [4]	10.7	-1.4	1.5	24.0
Li et al. [18]	9.2	-11.2	1.6	28.2
Triplanar FCN	6.7	3.5	0.7	6.5

Approach	VOE [%]	RVD [%]	ASD [mm]	MSSD [mm]
Li et al. [1]	14.4	8.1	2.4	7.2
Zikic et al. [19]	15.6	5.8	2.0	7.1
Triplanar FCN	13.3	6.3	1.8	7.5