北京生物医学工程

基于深度监督全卷积神经网络的 MRI脑图像语义分割算法

Semantic segmentation algorithm for MRI brain image based on deeply supervised fully convolutional network

作者：黄星奕丘子明许燕

单位：北京航空航天大学深圳研究院(北京 100091)

关键词：语义分割; 深度学习; 医学图像; 神经网络; 机器学习

分类号：R318.04

出版年·卷·期（页码）：2019·38·3（277-282）

摘要：

目的依据临床诊断对 MRI 脑图像自动分割算法的需求, 基于卷积神经网络(convolutional neural networks,CNN)设计了一种端到端的深度监督全卷积网络( deeply supervised fully convolutional network,DS-FCN)以解决脑图像中脑组织的自动分割问题? 方法针对三维 MRI 脑图像,先将体数据切割成二维图像切片,在 FCN 网络结构的基础上,加入了深度监督机制,即在特征提取的多层级结构中提前得到损失值反馈。结果以三维 MRI 脑图像公开数据集 LPBA-40 为实验数据,56 类脑组织的准确率(precision rate)，召回率(recall rate)，F1 评估值分别为74.40%，74.82%，73.75%，测试速率为 152 ms。结论通过引入深度监督结构,改进后的 DS-FCN 在 MRI 脑组织分割任务中得到了更精准的分割效果。

Objective In this paper, we present an end?to?end brain image semantic segmentation system based on the convolutional neural networks(CNN) framework to get segmentation of brain structures automatically,corresponding to the requirement of clinic automatic segmentation for MRI brain images ( DS-FCN,deeply supervised fully convolutional network). Methods For 3D MRI brain images,we firstly cut the volume data into 2D image slices.Then based on the FCN architecture,deep supervision is introduced into our system to get loss feedback in feature extraction across multiple scales. Results We select LONI,LPBA,40 as our experiment dataset which has 56 categories annotation of brain tissue. The precision,recall,F1,measure of our method reaches 74.40%,74.82% and 73.75%,which costs 152 ms ( on a typical GPU) to produce a segmentation map in testing phase. Conclusions Guided by the deeply multi?scale supervision, end?to?end segmentation system DS-FCN shows better results in experiment.

参考文献：

[ 1 ] Khayati R, Vafadust M, Towhidkhah F, et al. Fully automatic segmentation of multiple sclerosis lesions in brain MR FLAIR images using adaptive mixtures method and Markov random field model [ J]. Computers in Biology & Medicine, 2008,38 ( 3):

379-390.

[ 2 ] Jr CRJ,Bernstein MA,Fox NC,et al. The Alzheimer's Disease Neuroimaging Initiative ( ADNI): MRI Methods [ J]. Journal of Magnetic Resonance Imaging,2008,27(4):685-691.

[ 3 ] Xu Y,Jia Z,Ai Y,et al. Deep convolutional activation features for large scale Brain Tumor histopathology image classification and segmentation [ C ] / / IEEE. Brisbane, Australia: IEEE International Conference on Acoustics, Speech and Signal Processing,2015:947-951.

[ 4 ] Chen H,Wang X,Heng P A. Automated mitosis detection with deep regression networks [ C] / / IEEE, International Symposium on Biomedical Imaging. Prague: Czech Republic, 2016: 1204-1207.

[ 5 ] Hinton GE, Osindero S, Teh YW. A fast learning algorithm for deep belief nets [J]. Neural Computation, 2006, 18:1527-1554.

[ 6 ] Krizhevsky A, Sutskever I, Hinton GE. Imagenet classification with deep convolutional neural networks [ C ] / / Advances in neural information processing systems. Lake Tahoe,Nevada:MIT Press,2012: 1097-1105.

[ 7 ] Szegedy C,Liu W,Jia Y,et al. Going deeper with convolutions [C] / / Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Boston,MA:IEEE,2015: 1-9.

[ 8 ] He K,Zhang X,Ren S,et al. Deep residual learning for image recognition [ C ] / / Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas,NV:IEEE,2016: 770-778.

[ 9 ] Long J,Shelhamer E,Darrell T. Fully convolutional networks for semantic segmentation [ C ] / / Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Boston, MA:IEEE,2015: 3431-3440.

[10] Zheng S, Jayasumana S, Romera?Paredes B, et al. Conditional random fields as recurrent neural networks [ C] / / Proceedings of the IEEE International Conference on Computer Vision. Santiago, Chile:IEEE, 2015: 1529-1537.

[11] Lee CY, Xie S, Gallagher PW, et al. Deeply?Supervised Nets [ C ] / / The 18th International Conference on Artificial Intelligence and Statistics. San Diego, California : AISTATS.,2015: 562-570.

[12] Xie S, Tu Z. Holistically?nested edge detection [ C ] / /Proceedings of the IEEE International Conference on Computer Vision. Santiago,Chile:IEEE, 2015: 1395-1403.

[13] 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. Berlin: Springer International Publishing, 2015:234-241.

[14] Wu Z, Song S, Khosla A, et al. 3D shapenets: a deep representation for volumetric shapes [ C ] / / Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Boston,MA:IEEE,2015: 1912-1920.

[15] Shattuck DW,Mirza M,Adisetiyo V,et al. Construction of a 3D probabilistic atlas of human cortical structures[ J]. Neuroimage,2008,39(3):1064-1080.

服务与反馈：

【文章下载】【加入收藏】

提示：您还未登录，请登录！点此登录