ZHANG Jie,DENG Chengliang,XIE Shenghua,et al.Carotid Atherosclerosis Plaque Recognition Algorithm based on Deep Learning[J].Journal of Chengdu University of Information Technology,2021,36(02):143-148.[doi:10.16836/j.cnki.jcuit.2021.02.003]
基于深度学习的颈动脉粥样硬化斑块成分识别
- Title:
- Carotid Atherosclerosis Plaque Recognition Algorithm based on Deep Learning
- 文章编号:
- 2096-1618(2021)02-0143-06
- 关键词:
- Deeplab V3+; 图像分割; MobileNet; 颈动脉粥样硬化斑块; 成分识别
- Keywords:
- Deeplab V3+; image segmentation; MobileNet; carotid atherosclerotic plaque; composition analysis
- 分类号:
- TP301.6
- 文献标志码:
- A
- 摘要:
- 为实现颈动脉粥样硬化斑块成分的自动分割,提出一种基于深度学习的分割方法。针对颈动脉粥样硬化斑块成分特征复杂,医生手动分割提取费时且有误差等问题,使用基于Deeplab V3+网络的算法对颈动脉粥样硬化斑块成分进行自动分割来识别斑块成分。首先对已标识斑块成分的超声图像数据文件进行数据预处理、数据扩充以及感兴趣区域提取等操作,建立颈动脉粥样硬化斑块数据集,将数据集放到Deeplab V3+网络中训练和测试。针对Deeplab V3+网络复杂且参数量大的问题,在实验中结合MobileNet网络优点对原始网络进行优化。对比实验结果表明,优化后的模型在减少参数降低计算量的同时可以对颈动脉粥样硬化斑块成分进行有效分割与识别。
- Abstract:
- In order to achieve the automatic segmentation of carotid atherosclerotic plaque components, a deep learning based segmentation method was proposed. Firstly, the carotid atherosclerotic plaque data set was established through data preprocessing, data expansion and extraction of regions of interest, and the data set was trained and tested in Deeplab V3+ network. To solve the problem of Deeplab V3+ network with complex and large number of parameters, the original network was optimized by combining the advantages of MobileNet network in the experiment. According to the comparative experimental results, the optimized model can effectively segment and identify the components of carotid atherosclerosis plaque while reducing the amount of parameters and computation.
参考文献/References:
[1] 胡盛寿,高润霖,刘力生,等.《中国心血管病报告2018》概要[J].中国循环杂志,2019,34(3):209-220.
[2] 赵媛,孙夏,丁明跃.一种基于深度学习的颈动脉斑块超声图像识别方法[J].中国医疗器械信息,2017(9).
[3] Virmani R,Kolodgie F D,Burke A P,et al.Atherosclerotic Plaque Progression and Vulnerability to Rupture Angiogenesis as a Source of Intraplaque Hemorrhage[J].Arteriosclerosis Thrombosis & Vascular Biology,2005,25(10):2054.
[4] Gillard J,Graves M,Hatsukami T,et al.Carotid disease:the role of imaging in diagnosis and management[M].Cambridge University Press,2006.
[5] 罗俊,王淑珍.颈动脉粥样硬化与心血管疾病的关系[J].心血管病学进展,2011,32(2):236-238.
[6] 贾伟华,田桂玲.颈动脉斑块稳定性的研究进展[J].国际脑血管病杂志,2004(2):145-148.
[7] Zahalka A,Fenster A.An automated segmentation method for three-dimensional carotid ultrasound images[J].Physics in Medicine & Biology,2001,46(4):1321-1342.
[8] G E H.Reducing the dimensionality of data with neural networks[J].Science(New York,N.Y.),2006(313):5786.
[9] Hinton G E,Osindero S,Teh Y-W.A Fast Learning Algorithm for Deep Belief Nets[J].Neural Computation,2006,18(7):1527-1554.
[10] Menchón-Lara R M,Sancho-Gómez J L,Bueno-Crespo A.Early-stage atherosclerosis detection using deep learning over carotid ultrasound images[J].Applied Soft Computing,2016,49:616-628.
[11] 孙夏,吴蔚,吴鹏,等.基于卷积神经网络的颈动脉斑块超声图像特征识别[J].中国医疗器械信息,2016,22(9):4-8.
[12] Lu F,Wu F,Hu P,et al.Automatic 3D liver location and segmentation via convolutional neural network and graph cut[J].International Journal of Computer Assisted Radiology & Surgery,2017,12(2):171-182.
[13] Gao J,Wang Y-Y,Liu Y.Application of virtual histological intravascular ultrasound in plaque composition assessment of saphenous vein graft diseases[J].Chinese medical journal,2019,132(8):1.
[14] Chollet F.Xception:Deep Learning with Depthwise Separable Convolutions[J].IEEE Conference on Computer Vision and Pattern Recognition(CVPR),2016,195:1800-1807.
[15] Chen L C,Papandreou G,Schroff F,et al.Rethinking Atrous Convolution for Semantic Image Segmentation[J].arXiv,2017,17(6).
[16] 袁立,袁吉收,张德政.基于DeepLab-v3+的遥感影像分类[J].激光与光电子学进展,2019(15):236-243.
[17] Howard A G,Zhu M,Chen B,et al.MobileNets:Efficient Convolutional Neural Networks for Mobile Vision Applications[J].arXiv,2017,17(4):04861.
相似文献/References:
[1]唐明轩,李孝杰,周激流.基于Dense Connected深度卷积神经网络的
自动视网膜血管分割方法[J].成都信息工程大学学报,2018,(05):525.[doi:10.16836/j.cnki.jcuit.2018.05.007
]
TANG Ming-xuan,LI Xiao-jie,ZHOU Ji-liu.Automatic Retinal Vascular Segmentation Method based on
Densely Connected Convolution Neural Network[J].Journal of Chengdu University of Information Technology,2018,(02):525.[doi:10.16836/j.cnki.jcuit.2018.05.007
]
[2]杨 婞,邹书蓉,张洪伟.基于改进的量子行为粒子群聚类算法的彩色图像分割[J].成都信息工程大学学报,2016,(增刊1):14.
备注/Memo
收稿日期:2020-09-30
基金项目:四川省科技计划资助项目(2018JY0649)