多分类CNN的胶质母细胞瘤多模态MR图像分割

doi:10.3969/j.issn.0372-2112.2019.08.018

电子学报 ›› 2019, Vol. 47 ›› Issue (8): 1738-1747.DOI: 10.3969/j.issn.0372-2112.2019.08.018

所属专题：机器学习与智慧医疗

多分类CNN的胶质母细胞瘤多模态MR图像分割

赖小波¹, 许茂盛², 徐小媚¹

1. 浙江中医药大学医学技术学院, 浙江杭州 310053;
2. 浙江中医药大学第一临床医学院, 浙江杭州 310053

收稿日期:2018-08-30 修回日期:2019-03-17 出版日期:2019-08-25
- 作者简介:
- 赖小波 男,1981年生于江西赣州,博士,浙江中医药大学医学技术学院副教授,主要研究方向为数字医学影像处理与分析.E-mail:dmia_lab@zcmu.edu.cn;许茂盛 男,1966年生于浙江杭州,博士,浙江中医药大学第一临床医学院主任医师,主要研究方向为医学影像诊断学.E-mail:xms@sina.com
- 基金资助:
- 国家自然科学基金 (No.61602419）; 浙江省自然科学基金 (No.LY16F10008，No.LQ16F020003）

Glioblastoma Multiforme Multi-modal MR Images Segmentation Using Multi-class CNN

LAI Xiao-bo¹, XU Mao-sheng², XU Xiao-mei¹

1. Medical Technology College, Zhejiang Chinese Medical University, Hangzhou, Zhejiang 310053, China;
2. First Clinical Medicine College, Zhejiang Chinese Medical University, Hangzhou, Zhejiang 310053, China

Received:2018-08-30 Revised:2019-03-17 Online:2019-08-25 Published:2019-08-25
- Supported by:
- National Natural Science Foundation of China (No.61602419); National Natural Science Foundation of Zhejiang Province, China (No.LY16F10008, No.LQ16F020003)

摘要/Abstract

摘要： 为提高胶质母细胞瘤（GBM）多模态磁共振（MR）图像中各肿瘤子区域分割的准确性，提出一种多分类卷积神经网络（CNN）的GBM多模态MR图像自动分割算法.首先在98%缩尾处理和配准GBM多模态MR图像后，利用N4ITK法校正偏移场；其次构建一个主要由4个卷积层、2个池化层和2个全连接层组成的多分类CNN模型，训练后预分割GBM多模态MR图像，将体素分为5类不同的标签；最后移除所有小于200体素的假阳性区域，中值滤波后获得最终分割结果.以Dice相似性系数DSC、阳性预测值PPV和平均Hausdorff距离AHD为评价指标，利用所提出的算法对F-C-GBM数据集中整个肿瘤组织进行分割，获得的DSC、PPV、AHD分别为0.889±0.087、0.859±0.127和1.923.结果表明，该算法能有效提高GBM多模态MR图像分割的性能，可望有临床应用前景.

关键词: 胶质母细胞瘤, 多模态磁共振图像, 自动分割, 多分类卷积神经网络, 图像块

Abstract: To improve the accuracy of segmenting the tumor sub-regions in glioblastoma multiforme (GBM) multi-modal magnetic resonance (MR) images, a GBM multi-modal MR images automatic segmentation algorithm is proposed by using multi-class convolution neural network (CNN). Firstly, after 98% winsorization and registration for the GBM multi-modal MR images, the bias field was corrected by using the N4ITK method. Secondly, a multi-class CNN model mainly consisting of four convolutional layers, two pooling layers and two fully connected layers was constructed; the GBM multi-modal MR images were pre-segmented after training,and voxels were classified into five different labels.Finally,all false positive regions smaller than 200 voxels were removed,and the final segmentation results were obtained by median filtering. The Dice similarity coefficient DSC,positive predictive value PPV and average Hausdorff distance AHD were adopted as the evaluation index, and the DSC, PPV as well as AHD were 0.889±0.087, 0.859±0.127 and 1.923 for segmenting the entire tumor tissues in F-C-GBM dataset by the proposed algorithm, respectively. Results indicate that the proposed method can effectively improve the performance in the segmentation of the GBM multi-modal MR images and may be expected to have clinical application prospects.

Key words: glioblastoma multiforme, multi-modal magnetic resonance image, automatic segmentation, multi-class convolutional neural network, image patch

中图分类号:

TP391.41

赖小波, 许茂盛, 徐小媚. 多分类CNN的胶质母细胞瘤多模态MR图像分割[J]. 电子学报, 2019, 47(8): 1738-1747.

LAI Xiao-bo, XU Mao-sheng, XU Xiao-mei . Glioblastoma Multiforme Multi-modal MR Images Segmentation Using Multi-class CNN[J]. Acta Electronica Sinica, 2019, 47(8): 1738-1747.

参考文献

[1] 刘瑞.基于多模态MRI图像的脑肿瘤分割方法[D].成都:电子科技大学,2017. LIU Rui.Brain tumor segmentation methods based on multi-modal MRI images[D].Chengdu:University of Electronic Science and Technology,2017.(in Chinese)
[2] AKKUS Z,GALIMZIANOVA A,HOOGI A,et al.Deep learning for brain MRI segmentation:state of the art and future directions[J].Journal of Digital Imaging,2017,30(4):449-459.
[3] STIJN B,INGEBORG G,ROEL V H.Machine learning based brain tumour segmentation on limited data using local texture and abnormality[J].Computers in Biology and Medicine,2018,98(1):39-47.
[4] KAUR T,SAINI B S,GUPTA S.A novel fully automatic multilevel thresholding technique based on optimized intuitionistic fuzzy sets and tsallis entropy for MR brain tumor image segmentation[J].Australasian Physical & Engineering Sciences in Medicine,2018,41(1):41-58.
[5] VISHNUVARTHANAN G,RAJASEKARAN M P,VISHNUVARTHANAN N A,et al.Tumor detection in T1,T2,FLAIR and MPR brain images using a combination of optimization and fuzzy clustering improved by seed-based region growing algorithm[J].International Journal of Imaging Systems and Technology,2017,27(1):33-45.
[6] KAVITHA A R,CHELLAMUTHU C.Brain tumour segmentation from MRI image using genetic algorithm with fuzzy initialisation and seeded modified region growing(GFSMRG) method[J].Imaging Science Journal,2016,64(5):285-297.
[7] SASIKANTH S,KUMAR S S.Glioma tumor detection in brain MRI image using ANFIS-based normalized graph cut approach[J].International Journal of Imaging Systems and Technology,2018,28(1):64-71.
[8] VARUNA S N,KUMAR T N R.Identification and classification of brain tumor MRI images with feature extraction using DWT and probabilistic neural network[J].Brain Informatics,2018,5(1):23-30.
[9] ESSADIKE A,OUABIDA E,BOUZID A.Brain tumor segmentation with Vander Lugt correlator based active contour[J].Computer Methods and Programs in Biomedicine,2018,160(9):103-117.
[10] LOK K H,SHI L,ZHU X L,et al.Fast and robust brain tumor segmentation using level set method with multiple image information[J].Journal of X-Ray Science and Technology,2017,25(2):301-312.
[11] KARUPPATHAL R,PALANISAMY V.An automotive approach for brain tumor segmentation based on Gaussian distribution and level set method[J].Current Medical Imaging Reviews,2014,10(4):290-296.
[12] POPE W B,SAYRE J,PERLINA A,et al.MR imaging correlates of survival in patients with high-grade gliomas[J].American Journal of Neuroradiology,2005,26(10):2466-2474.
[13] 柯圣财,赵永威,李弼程,等.基于卷积神经网络和监督核哈希的图像检索方法[J].电子学报,2017,45(1):158-163. KE Sheng-cai,ZHAO Yong-wei,LI Bi-cheng,et al.Image retrieval based on convolutional neural network and kernel-based supervised hashing[J].Acta Electronica Sinica,2017,45(1):158-163.(in Chinese)
[14] JAYASURIYA S A,LIEW A W C,LAW N F.Symmetry plane detection in brain image analysis:a survey[J].Current Medical Imaging Reviews,2013,9(3):230-247.
[15] TUSTISON N J,AVANTS B B,COOK P A,et al.N4ITK:improved N3 bias correction[J].IEEE Transactions on Medical Imaging,2010,29(6):1310-1320.
[16] PEREIRA S,PINTO A,ALVES V,et al.Brain tumor segmentation using convolutional neural networks in MRI images[J].IEEE Transactions on Medical Imaging,2016,30(5):1240-1251.
[17] HAVAEI M,DAVY A,WARDE-FARLEY D,et al.Brain tumor segmentation with deep neural networks[J].Medical Image Analysis,2017,35:18-31.
[18] ZIKIC D,GLOCKER B,KONUKOGLU E,et al.Decision forests for tissue-specific segmentation of high-grade gliomas in multi-channel MR[A].International Conference on Medical Image Computing and Computer-Assisted Intervention[C].Heidelberg,Berlin:Springer,2012.369-376.
[19] MEIER R,BAUER S,SLOTBOOM J,et al.Appearance-and context-sensitive features for brain tumor segmentation[A].International Conference on Medical Image Computing and Computer-Assisted Intervention[C].Heidelberg,Berlin:Springer,2015.48-51.
[20] REZA S,IFTEKHARUDDIN K.Multi-fractal texture features for brain tumor and edema segmentation[A].International Society for Optics and Photonics[C].Bellingham,Washington:SPIE,2014.903503-903503.
[21] SUBBANNA N,PRECUP D,COLLINS L,et al.Hierarchical probabilistic gabor and MRF segmentation of brain tumours in MRI volumes[A].International Conference on Medical Image Computing and Computer-Assisted Intervention[C].Heidelberg,Berlin:Springer,2013.751-758.
[22] TUSTISON N J,SHRINIDHI K L,WINTERMARK M,et al.Optimal symmetric multimodal templates and concatenated random forests for supervised brain tumor segmentation (simplified) with ANTsR[J].Neuroinformatics,2015,13(2):209-225.
[23] KWON D,SHINOHARA R T,AKBARI H,et al.Combining generative models for multifocal glioma segmentation and registration[A].International Conference on Medical Image Computing and Computer-Assisted Intervention[C].Heidelberg,Berlin:Springer,2014.763-770.

多分类CNN的胶质母细胞瘤多模态MR图像分割

Glioblastoma Multiforme Multi-modal MR Images Segmentation Using Multi-class CNN

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 7

编辑推荐

Metrics

[1]	赖小波, 许茂盛, 徐小媚. 基于多模型融合和区域迭代生长的视网膜血管自动分割[J]. 电子学报, 2019, 47(12): 2611-2621.
[2]	彭圆圆, 肖昌炎. 一种全自动的肺裂分割方法[J]. 电子学报, 2018, 46(6): 1319-1326.
[3]	赵峙江;张田文;张志宏. 一种新的基于PCNN的图像自动分割算法研究[J]. 电子学报, 2005, 33(7): 1342-1344.
[4]	毕英伟, 邱天爽. 一种基于简化PCNN的自适应图像分割方法[J]. 电子学报, 2005, 33(4): 647-650.
[5]	秦磊;舒华忠;於文雪;金丰华;C.Toumoulin;罗立民. Legendre矩的两种快速算法[J]. 电子学报, 2004, 32(1): 25-28.
[6]	陈韩锋, 戚飞虎. 结合多种语义信息的半自动视频对象分割[J]. 电子学报, 2002, 30(S1): 2012-2015.
[7]	蔡兴建;曹毅;毛军发;李征帆. 高速双面共面结构印刷电路板电特性仿真[J]. 电子学报, 2001, 29(2): 287-288.