基于自适应多分辨率特征学习的CNV分型网络

doi:10.6040/j.issn.1672-3961.0.2021.604

摘要/Abstract

摘要： 为解决不同脉络膜新生血管(choroidal neovascularization, CNV)类型间较小区分性带来的分型难度和光学相干断层扫描(optical coherence tomography, OCT)图像中噪声对分型精度的影响,提出自适应多分辨率特征学习的CNV分型方法,其包含多分辨率特征学习和自适应特征选择模块。在多分辨率特征学习模块中,融合具有不同类型CNV细节信息的底层特征和具有语义信息的高层特征,同时引入渐进式的训练方式增强特征表示能力。在自适应特征选择模块中,通过引入注意力机制,对最后分型起关键作用的特征进行增强,进一步提升特征的区分性。在自建的CNV数据集上进行试验,试验结果表明,评价指标上的测试评分分别为91.3%、86.6%、89.2%和90.6%。提出的自适应多分辨率特征学习的CNV分型方法优于现有的其他分类方法。

关键词: CNV分型, 自适应多分辨率特征学习, 光学相干断层扫描图像, 自适应特征选择, 多分辨率特征学习

中图分类号:

TP391.4

许传臻,袭肖明,李维翠,孙仪,杨璐. 基于自适应多分辨率特征学习的CNV分型网络[J]. 山东大学学报 (工学版), 2022, 52(4): 69-75.

XU Chuanzhen, XI Xiaoming, LI Weicui, SUN Yi, YANG Lu. Adaptive multi-resolution feature learning network for CNV classification[J]. Journal of Shandong University(Engineering Science), 2022, 52(4): 69-75.

参考文献

[1] XI Xiaoming, MENG Xianjing, QIN Zheyun, et al. IA-net: informative attention convolutional neural network for choroidal neovascularization segmentation in OCT images[J]. Biomedical Optics Express, 2020,11(11):6122-6136.
[2] SERRA R, COSCAS F, PINNA A, et al. Quantitative optical coherence tomography angiography features of inactive macularneovascularization in age-related macular degeneration[J]. Retina, 2021, 41(1):93-102.
[3] ZHU Shuxia, SHI Fei, XIANG Dehui, et al. Choroid neovascularization growth prediction with treatment based on reaction-diffusion model in 3-D OCT images[J]. IEEE Journal of Biomedical and Health Informatics, 2017, 21(6):1667-1674.
[4] WANG Bo, WEI Wei, QIU Shuang, et al. Boundary aware u-net for retinal layers segmentation in optical coherence tomography images[J]. IEEE Journal of Biomedical and Health Informatics, 2021, 25(8):3029-3040.
[5] ZHANG Huihong, YANG Jianlong, ZHOU Kang, et al. Automatic segmentation and visualization of choroid in OCT with knowledge infused deep learning[J]. IEEE Journal of Biomedical and Health Informatics, 2020, 24(12):3408-3420.
[6] SULZBACHER F, POLLREISZ A, KAIDER A, et al. Identification and clinical role ofchoroidal neovas-cularization characteristics based on optical coherence tomography angiography[J]. Acta Ophthalmol, 2017, 95(4):414-420.
[7] HUANG G, LIU Z, VAN L, et al. Densely connected convolutional networks[C] //Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Salt Lake City, USA: IEEE, 2018:4700-4708.
[8] KH I, SUDANTHI W, STEPHEN O. Identifying diabetic retinopathy fromoct images using deep transfer learning with artificial neural networks[C] //Proceedings of the 2019 IEEE 32nd International Symposium on Computer-Based Medical Systems(CBMS). Córdoba,Spain:CBMS, 2019:281-286.
[9] FANG Leyuan, JIN Yuxuan, HUANG Laifeng, et al. Iterative fusion convolutional neural networks for classification of optical coherence tomography images[J]. Journal of Visual Communication and Image Representation, 2019, 59:327-333.
[10] VINEETA D, EEDARA P, SAMARENDRA D, et al. B-scan attentive CNN for the classification of retinal optical coherence tomography volumes[J]. IEEE Signal Processing Letters, 2020, 27:1025-1029.
[11] KRAUSE J, STARK M, DENG Jia, et al. 3d object representations for fine-grained categorization[C] //Proceedings of the IEEE International Conference on Computer Vision Workshops.Portland, Australia: ICCV, 2013:554-561.
[12] MAJI S, RAHTU E, KANNALA J, et al. Fine-grained visual classification of aircraft[J]. ArXiv, 2013, 25(9):1306-1312.
[13] CHANG Dongliang, DING Yifeng, XIE Jiyang, et al. The devil is in the channels: mutual-channel loss for fine-grained image classification[J]. IEEE Transactions on Image Processing, 2020, 29:4683-4695.
[14] ZHANG Ziyue, JI Zexian, CHEN Jiang, et al. Joint optimization of cycleGAN and CNN classifier for detection and localization of retinal pathologies on color fundus photographs[J]. IEEE Journal of Biomedical and Health Informatics, 2021, 26(1):115-126.
[15] BRANSON S, GRANT V, SERGE B, et al. Bird species categorization using pose normalized deepconvolutional nets[J]. ArXiv, 2014, 24(8):1406-1420.
[16] WANG Y, MORARIU V, Larry S. Learning a discriminative filter bank within a cnn for fine-grained recognition[C] //Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Salt Lake City, USA: CVPR, 2018:4148-4157.
[17] LIN T, ARUNI R, SUNHRANSU M. Bilinear CNN models forfine-grained visual recognition[C] //Proceedings of the IEEE International Conference on Computer Vision. Santiago, Chile: ICCV, 2015:1449-1457.
[18] CUI Yin, ZHOU Feng, WANG Jiang, et al. Kernel pooling for convolutional neural networks[C] //Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Honolulu, USA: CVPR, 2017:2921-2930.
[19] CAI Sijia, ZUO Wangmeng, ZHANG Lei. Higher-order integration of hierarchical convolutional activations for fine-grained visual categorization[C] //Proceedings of the IEEE International Conference on Computer Vision. Venice, Italy: ICCV, 2017:511-520.
[20] HE Kaiming, ZHANG Xiangyu, REN Shaoqing, et al. Deep residual learning for image recognition[C] //Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, USA: CVPR, 2016:770-778.
[21] 白琮,黄玲,陈佳楠,等.面向大规模图像分类的深度卷积神经网络优化[J].软件学报,2018,29(4):1029-1038. BAI Zong, HUANG Ling, CHEN Jianan, et al. Optimization of deep convolution neural network for large-scale image classification[J]. Journal of Software, 2018, 29(4):1029-1038.
[22] 庞浩,王枞.用于糖尿病视网膜病变检测的深度学习模型[J].软件学报,2017,28(11):3018-3029. PANG Hao, WANG Cong. Deep learning model for diabetic retinopathy detection[J]. Journal of Software, 2017, 28(11):3018-3029.
[23] JIAN M, LAM K. Simultaneous hallucination and recognition of low-resolution faces based on singular value decomposition[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2015, 25(11):1761-1772.
[24] JIAN M, LAM K, DONG J. Facial-feature detection and localization based on a hierarchical scheme[J]. Information Sciences, 2014, 262:1-14.
[25] 艾星芳,谢鑫鹏.基于多分辨率融合深度学习网络的视网膜病变分割[J].现代计算机,2021(5):52-56. AI Xingfang, XIE Xingpeng. Retinopathy segmentation based on multi-resolution fusion depth learning network[J]. Modern Computer, 2021(5):52-56.
[26] YANG Ze, LUO Tiange, WANG Dong, et al. Learning to navigate for fine-grained classification[C] //Proceedings of the European Conference on Computer Vision(ECCV). Munich, Germany: ECCV, 2018:420-435.
[27] ZHENG Heliang, FU Jianlong, ZHA Zhengjun, et al. Looking for the devil in the details: Learning trilinear attention sampling network for fine-grained image recognition[C] //Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Long Beach, USA: CVPR, 2019:1903-1913.
[28] SELVARAJU R, MICHAEL C, ABHISHEK D, et al. Grad-cam:visual explanations from deep networks via gradient-based localization[C] //Proceedings of the IEEE International Conference on Computer Vision. Venice, Italy: ICCV, 2017:618-626.

相关文章 15

[1]	梁晔,马楠,刘宏哲. 图像依赖的显著图融合方法[J]. 山东大学学报 (工学版), 2021, 51(4): 1-7.
[2]	周风余,顾潘龙,万方,尹磊,贺家凯. 多运动视觉里程计的方法与技术[J]. 山东大学学报 (工学版), 2021, 51(1): 1-10.
[3]	刘保成,朴燕,宋雪梅. 联合检测的自适应融合目标跟踪[J]. 山东大学学报 (工学版), 2020, 50(3): 51-57.
[4]	张胜男,王雷,常春红,郝本利. 基于三维剪切波变换和BM4D的图像去噪方法[J]. 山东大学学报 (工学版), 2020, 50(2): 83-90.
[5]	赵越男,陈桂友,孙琛,卢宁,廖立伟. 基于空间隐患分布与运动意图解析的危险评估方法[J]. 山东大学学报 (工学版), 2020, 50(1): 28-34.
[6]	姚元玺. 基于分场景重构的风电汇聚趋势性量化方法[J]. 山东大学学报 (工学版), 2019, 49(6): 86-92.
[7]	汪嘉晨,唐向红,陆见光. 轴承故障诊断中特征选取技术[J]. 山东大学学报 (工学版), 2019, 49(2): 80-87, 95.
[8]	刘世光,王海荣,刘锦. 快速四点一致性点云粗配准算法[J]. 山东大学学报 (工学版), 2019, 49(2): 1-7.
[9]	范君,业巧林,业宁. 基于改进的有监督无参局部保持投影算法的人脸识别[J]. 山东大学学报 (工学版), 2019, 49(1): 10-16.
[10]	王国新,陈凤东,刘国栋. 基于彩色伪随机编码结构光特征提取方法[J]. 山东大学学报 (工学版), 2018, 48(5): 55-60.
[11]	窦婷婷,姚元玺,陈鹏,芦灯. 基于ATP-EMTP的电弧建模及工程仿真[J]. 山东大学学报(工学版), 2018, 48(4): 102-108.
[12]	张宪红,张春蕊. 基于六维前馈神经网络模型的图像增强算法[J]. 山东大学学报(工学版), 2018, 48(4): 10-19.
[13]	丁筱玲,赵强,李贻斌,马昕. 基于模板匹配的改进型目标识别算法[J]. 山东大学学报(工学版), 2018, 48(2): 1-7.
[14]	赵晔,何潇,周东华. 带有不准确测量噪声的最小二乘故障估计[J]. 山东大学学报(工学版), 2017, 47(5): 254-262.
[15]	刘岩,李幼军,陈萌. 基于EMD和SVM的抑郁症静息态脑电信号分类研究[J]. 山东大学学报(工学版), 2017, 47(3): 21-26.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed