Automatic Precise Segmentation of Cerebellopontine Angle Tumor Based on Faster-RCNN and Level-Set Method

doi:10.11938/cjmr20212881

Abstract

Abstract:

To meet the demands in surgical treatment and radiotherapy, this work combines the faster region convolutional neural network (Faster-RCNN) and Level-Set methods to segment cerebellopontine angle (CPA) tumors automatically and precisely. T₁WI-SE magnetic resonance images from 317 CPA tumor patients were collected. Features extracted by VGG16 were combined with the region proposal network (RPN) for training. A CPA tumor localization model was then established, before the Level-Set method was applied to accurately segment the tumor. The segmentation results of different CPA tumor regions were compared in terms of precision, recall, mean average precision (mAP) and Dice coefficient. The results showed that the method proposed can effectively and precisely segment CPA tumors, thereby capable of reducing the burden on clinicians and improving the treatment effect.

Key words: faster region convolutional neural network (Faster-RCNN), Level-Set, image segmentation, cerebellopontine angle tumor

CLC Number:

O482.53

Ying LIU,Yi-yun GUO,Jing-cong CHEN,Hao-wei ZHANG. Automatic Precise Segmentation of Cerebellopontine Angle Tumor Based on Faster-RCNN and Level-Set Method[J]. Chinese Journal of Magnetic Resonance, 2021, 38(3): 381-391.

Figures/Tables 9

Fig.1

Fig.2

Fig.3

Fig.4

Fig.5

Fig.6

Fig.7

Table 1

Table 2

References 29

1	PING X X , MENG Q , TIAN X , et al. MRI findings of lesions in the cerebellopontine angle[J]. J Med Imaging, 2014, 24 (1): 12- 15. doi: 10.3969/j.issn.1005-5185.2014.01.004
	平小夏, 孟倩, 田霞, 等. 桥小脑角区病变的MRI表现[J]. 医学影像学杂志, 2014, 24 (1): 12- 15. doi: 10.3969/j.issn.1005-5185.2014.01.004
2	FANG F , HU S P . Comparative analysis of MRI findings and pathology in acoustic neuroma[J]. CT Theory and Applications, 2019, 28 (6): 731- 738.
	方芳, 胡少平. 听神经瘤MRI表现与病理对照分析[J]. CT理论与应用研究, 2019, 28 (6): 731- 738.
3	CHEN Q , LI G Q , LI J T . Study of the value MRI in the diagnosis of intracranial meningioma[J]. Chinese Jopurnal of CT and MRI, 2016, 14 (4): 23- 26. doi: 10.3969/j.issn.1672-5131.2016.04.008
	陈琪, 李国强, 李惊涛. MRI对颅内脑膜瘤的诊断价值研究[J]. 中国CT和MRI杂志, 2016, 14 (4): 23- 26. doi: 10.3969/j.issn.1672-5131.2016.04.008
4	ZHANG J Q , ZHANG Y , YIN Y , et al. A review of machine learning in tumor radio therapy[J]. Journal of Biomedical Engineering, 2019, 36 (5): 879- 844.
	张珺倩, 张远, 尹勇, 等. 机器学习在肿瘤放射治疗领域应用进展[J]. 生物医学工程学杂志, 2019, 36 (5): 879- 884.
5	LI Q , BAI K X , ZHAO L , et al. Progresss and challenges of MRI brain tumor image segmentation[J]. Journal of Image and Graphics, 2020, 25 (3): 419- 431.
	李锵, 白柯鑫, 赵柳, 等. MRI脑肿瘤图像分割研究进展及挑战[J]. 中国图象图形学报, 2020, 25 (3): 419- 431.
6	WANG C F , WANG B J , WANG Z , et al. The target profile of multimodal magnetic resonance imaging in the radiotherapy of craniocerebral malignant tumor[J]. Chinese Journal of Practical Nervous Diseases, 2018, 21 (19): 2118- 2124. doi: 10.12083/SYSJ.2018.19.462
	王长福, 王斌杰, 王智, 等. 多模态磁共振成像及融合技术在颅脑恶性肿瘤放疗靶区勾画中的应用[J]. 中国实用神经疾病杂志, 2018, 21 (19): 2118- 2124. doi: 10.12083/SYSJ.2018.19.462
7	ZIKIC D, IOANNOU Y, BROWN M, et al. Segmentation of brain tumor tissues with convolutional neural networks[C]. Proceedings MICCAI-BRATS, 2014: 36-39.
8	JIA Y, SHELHAMER E, DONAHUE J, et al. Caffe: Convolutional architecture for fast feature embedding[C]//Proceedings of the 22nd ACM international conference on Multimedia. 2014: 675-678.
9	IQBAL S , GHANI M U , SABA T , et al. Brain tumor segmentation in multi-spectral MRI using convolutional neural networks (CNN)[J]. Microsc Res Techniq, 2018, 81 (4): 419- 427. doi: 10.1002/jemt.22994
10	EZHILARASI R, VARALAKSHMI P. Tumor detection in the brain using faster R-CNN[C]//20182nd International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud). IEEE, 2018: 388-392.
11	YANG X H , ZHANG Y . Brain tumor segmentation based on MRI multimodal information and 3D-CNNs feature extraction[J]. Chinese Journal of CT and MRI, 2020, 18 (9): 4- 6, 23. doi: 10.3969/j.issn.1672-5131.2020.09.002
	杨新焕, 张勇. 结合MRI多模态信息和3D-CNNs特征提取的脑肿瘤分割研究[J]. 中国CT和MRI杂志, 2020, 18 (9): 4- 6, 23. doi: 10.3969/j.issn.1672-5131.2020.09.002
12	THILLAIKKARASI R , SARAVANAN S . An enhancement of deep learning algorithm for brain tumor segmentation using kernel based CNN with M-SVM[J]. J Med Syst, 2019, 43 (4): 1- 7. doi: 10.1007/s10916-019-1223-7
13	IQBAL S , GHANI KHAN M U , SABA T , et al. Deep learning model integrating features and novel classifiers fusion for brain tumor segmentation[J]. Microsc Res Techniq, 2019, 82 (8): 1302- 1315. doi: 10.1002/jemt.23281
14	KASS M , WITKIN A , TERZOPOULOS D . Snakes: Active contour models[J]. Int J Comput Vision, 1988, 1 (4): 321- 331. doi: 10.1007/BF00133570
15	CHAN T F , VESE L A . Active contours without edges[J]. IEEE T Image Process, 2001, 10 (2): 266- 277. doi: 10.1109/83.902291
16	LI C M , KAO C Y , GORE J C , et al. Minimization of region-scalable fitting energy for image segmentation[J]. IEEE T Image Process, 2008, 17 (10): 1940- 1949. doi: 10.1109/TIP.2008.2002304
17	YIN S L , LI H , LIU D S , et al. Active contour modal based on density-oriented BIRCH clustering method for medical image segmentation[J]. Multimed Tools Appl, 2020, 79 (41): 31049- 31068.
18	YUAN C A , ZHENG Y , QIN X , et al. The medical image segmentation method of improved RSF active contour model[J]. Journal of Zhengzhou University (Natural Science Edition), 2017, 49 (1): 34- 38, 44.
	元昌安, 郑彦, 覃晓, 等. 改进RSF主动轮廓模型的医学图像分割方法[J]. 郑州大学学报(理学版), 2017, 49 (1): 34- 38.
19	SHI Q Y , YAN F , YANG Y , et al. Image segmentation of tooth and alveolar bone with the level set model[J]. Chinese J Magn Reson, 2021, 38 (2): 182- 193.
	石沁祎, 闫方, 杨阳, 等. 基于水平集的牙齿牙槽骨图像分割[J]. 波谱学杂志, 2021, 38 (2): 182- 193.
20	ZHANG D, ZHU W, ZHAO H, et al. Automatic localization and segmentation of optical disk based on faster R-CNN and level set in fundus image[C]//Medical Imaging 2018: Image Processing. International Society for Optics and Photonics, 2018, 10574: 105741U.
21	ZUO J H , ZHAO C , ZHU X L , et al. High-resolution remote sensing image building extraction combined with Faster-RCNN and Level-Set[J]. Chinese Journal of Liquid Crystals and Displays, 2019, 34 (4): 439- 447.
	左俊皓, 赵聪, 朱晓龙, 等. Faster-RCNN和Level-Set结合的高分遥感影像建筑物提取[J]. 液晶与显示, 2019, 34 (4): 439- 447.
22	LIU Y , CHEN J C , HU X Y , et al. Classification and localization of acoustic neuroma and meningioma in the cerebellopontine angle based on Mask RCNN[J]. Chinese J Magn Reson, 2021, 38 (1): 58- 68.
	刘颖, 陈静聪, 胡小洋, 等. 基于Mask RCNN的桥小脑角区听神经瘤与脑膜瘤分类定位研究[J]. 波谱学杂志, 2021, 38 (1): 58- 68.
23	REN S Q, HE K M, GIRSHICK R, et al. Faster R-CNN: Towards real-time object detection with region proposal networks[C]//Advances in neural information processing systems. 2015: 91-99.
24	OSHER S , SETHIAN J A . Fronts propagating with curvature-dependent speed: algorithms based on Hamilton-Jacobi formulations[J]. J Comput Phys, 1988, 79 (1): 12- 49. doi: 10.1016/0021-9991(88)90002-2
25	YAO C W , HUANG D B , YE M Q . MR brain tumor image segmentation based on rough set adaptive granularity[J]. Computer and Modernization, 2021, (1): 34- 37. doi: 10.3969/j.issn.1006-2475.2021.01.007
	姚传文, 黄道斌, 叶明全. 基于粗糙集自适应粒度的MR脑肿瘤图像分割[J]. 计算机与现代化, 2021, (1): 34- 37. doi: 10.3969/j.issn.1006-2475.2021.01.007
26	THAHA M M , KUMAR K P M , MURUGAN B S , et al. Brain tumor segmentation using convolutional neural networks in MRI images[J]. J Med Syst, 2019, 43 (9): 1- 10.
27	GE T , ZHAN T M , MU S X . Brain tumor segmentation algorithm based on multi-kernel collaborative representation classification[J]. Journal of Nanjing University of Science and Technology, 2019, 43 (5): 578- 585.
	葛婷, 詹天明, 牟善祥. 基于多核协同表示分类的脑肿瘤分割算法[J]. 南京理工大学学报, 2019, 43 (5): 578- 585.
28	RONNEBERGER O, FISCHER P, BROX T. U-net: Convolutional networks for biomedical image segmentation[C]//International Conference on Medical image computing and computer-assisted intervention. Springer, Cham, 2015: 234-241.
29	XU Z W, WU Z Y, FENG J J. CFUN: combining faster R-CNN and U-net network for efficient whole heart segmentation[J]. arXiv preprint arXiv: 1812.04914, 2018.

	Loss	Precision	Recall	mAP	Dice
勾画范围扩大前	0.0511	0.9675	0.4990	0.8063	0.8670
勾画范围扩大1.5倍	0.0136	0.9210	0.6412	0.8994	0.8887
勾画范围扩大2倍	0.0938	0.8743	0.6733	0.7842	0.7935

分割方法	Dice
粗糙集自适应粒度^[25]	0.7734
ECNN^[26]	0.8534
多核协同^[27]	0.8700
Unet^[28]	0.8234
CFUN^[29]	0.8775
本文方法	0.8887

[1]	Qin-yi SHI,Fang YAN,Yang YANG,Yue-fu CHEN,Xiao-lang LIN,Yuan-jun WANG. Image Segmentation of Tooth and Alveolar Bone with the Level Set Model [J]. Chinese Journal of Magnetic Resonance, 2021, 38(2): 182-193.
[2]	GONG Jin-chang, WANG Yu, WANG Yuan-jun. A Method for Segmentation of Glioma on Multimodal Magnetic Resonance Images Based on Wavelet Fusion and Deep Learning [J]. Chinese Journal of Magnetic Resonance, 2020, 37(2): 131-143.
[3]	XIAO Liang, LOU Yu-kun, ZHOU Hang-yu. A U-Net Network-Based Rapid Construction of Knee Models for Specific Absorption Rate Estimation [J]. Chinese Journal of Magnetic Resonance, 2020, 37(2): 144-151.
[4]	ZHANG Xue-ying, WANG Cheng-long, XIE Hai-bin, ZHANG Cheng-xiu, MA Chao, LU Jian-ping, YANG Guang. Parallel Segmentation and Tracking Algorithm for Magnetic Resonance Angiography Images Based on GPU [J]. Chinese Journal of Magnetic Resonance, 2016, 33(4): 570-580.