权重初始化-滑动窗口CNN的医学图像分类

doi:10.3778/j.issn.1673-9418.2011091

计算机科学与探索 ›› 2022, Vol. 16 ›› Issue (8): 1885-1897.DOI: 10.3778/j.issn.1673-9418.2011091

权重初始化-滑动窗口CNN的医学图像分类

安凤平¹^,⁺(), 李晓薇¹, 曹翔¹

1. 淮阴师范学院物理与电子电气工程学院，江苏淮安 223300
2. 北京理工大学信息与电子学院，北京 100081

收稿日期:2020-11-27 修回日期:2021-01-29 出版日期:2022-08-01 发布日期:2021-03-03
通讯作者: +E-mail: anfengping@163.com。
作者简介:安凤平（1985—），男，安徽芜湖人，博士，副教授，硕士生导师，主要研究方向为图像处理、深度学习。
李晓薇（1962—），女，江苏扬州人，硕士，教授，硕士生导师，主要研究方向为超导隧道结、混杂系统。
曹翔（1981—），男，四川荣县人，博士，副教授，硕士生导师，主要研究方向为多水下机器人协作控制、水下机器人任务规划。
基金资助:
国家自然科学基金(61701188);江苏省自然科学基金(BK20201479);中国博士后科学基金(2019M650512)

Medical Image Classification Algorithm Based on Weight Initialization-Sliding Window CNN

AN Fengping¹^,⁺(), LI Xiaowei¹, CAO Xiang¹

1. School of Physics and Electronic Electrical Engineering, Huaiyin Normal University, Huai’an, Jiangsu 223300, China
2. School of Information and Electronics, Beijing Institute of Technology, Beijing 100081, China

Received:2020-11-27 Revised:2021-01-29 Online:2022-08-01 Published:2021-03-03
About author:AN Fengping, born in 1985, Ph.D., associate professor, M.S. supervisor. His research interests include image processing and deep learning.
LI Xiaowei, born in 1962, M.S., professor, M.S. supervisor. Her research interests include superconducting tunnel junction and hybrid system.
CAO Xiang, born in 1981, Ph.D., associate professor, M.S. supervisor. His research interests include cooperative control of multiple underwater robots and task planning of underwater robots.
Supported by:
This work was supported by the National Natural Science Foundation of China(61701188);the Natural Science Foundation of Jiangsu Province(BK20201479);the Postdoctoral Science-Foundation of China(2019M650512)

摘要/Abstract

摘要：

深度学习在医学图像分类应用过程中存在以下问题：一是无法针对医学图像性质构建深度学习模型层级;二是深度学习模型网络初始化权重未能得到较好优化。为此，首先从网络优化角度出发，通过优化方法提高网络的非线性建模能力，提出了一种新的网络权重初始化方法，缓解了现有深度学习的初始化理论受限于非线性单元类型的问题，增加了神经网络处理不同视觉任务的潜力。同时，为了充分利用医学图像的特性，通过对多列卷积神经网络框架进行深入研究，发现通过改变卷积神经网络不同层次的特征数目和卷积核大小，可以构建不同的卷积神经网络模型，以更好地适应待处理医学图像的医学特性，并训练得到的异构多列卷积神经网络。最后，利用提出的一种自适应的滑动窗口融合机制，共同完成医学图像的分类任务。基于上述思想，提出了一种基于权重初始化-多层卷积神经网络滑动窗口融合的医学分类算法。利用提出方法对乳腺肿块分类、脑肿瘤组织分类实验和医学图像数据库分类分别进行实验，实验结果表明，所提方法不仅平均准确率较传统机器学习、其他深度学习方法有明显提高，而且具有较好的稳定性和鲁棒性。

关键词: 深度学习, 权重初始化, 滑动窗口, 卷积神经网络（CNN）, 医学图像分类

Abstract:

Deep learning has the following problems in medical image classification application: first, it is impossible to construct a deep learning model hierarchy for medical image properties; second, the network initialization weight of the deep learning model has not been optimized. To this end, this paper starts from the perspective of network optimization, and then improves the nonlinear modeling ability of the network through optimization methods. Then, this paper proposes a new network weight initialization method, which alleviates the problem that the initialization theory of existing deep learning is limited by the nonlinear unit type, and increases the potential of neural network to deal with different visual tasks. At the same time, in order to make full use of the characteristics of medical images, this paper deeply studies the multi-column convolutional neural network framework and finds that through changing the number of features and the convolution kernel size of different levels of convolutional neural networks, it can construct different convolutional neural network models to better adapt to the medical characteristics of the medical images to be processed and train the obtained heterogeneous multi-column convolutional neural networks. Finally, the classification task of medical images is completed by the method proposed in this paper. Based on the above ideas, this paper proposes a medical classification algorithm based on weight initialization-sliding window fusion of multi-layer convolutional neural networks. The methods of this paper are used to classify breast mass classification, brain tumor tissue classification experiment and medical image database classification. The experimental results show that the proposed method not only has higher average accuracy than traditional machine learning and other deep learning methods, but also has better stability and robustness.

Key words: deep learning, weight initialization, sliding window, convolutional neural network (CNN), medical image classification

中图分类号:

TP751

安凤平, 李晓薇, 曹翔. 权重初始化-滑动窗口CNN的医学图像分类[J]. 计算机科学与探索, 2022, 16(8): 1885-1897.

AN Fengping, LI Xiaowei, CAO Xiang. Medical Image Classification Algorithm Based on Weight Initialization-Sliding Window CNN[J]. Journal of Frontiers of Computer Science and Technology, 2022, 16(8): 1885-1897.

图/表 11

图1 本文所提的多层卷积神经网络结构图

Fig.1 Structure diagram of multi-layer convolutional neural network proposed in this paper

图2 多层卷积神经网络滑动窗口融合示意图

Fig.2 Schematic diagram of sliding window fusion of multi-layer convolutional neural network

表1 滑动窗口融合方法的特例

Table 1 Special cases of sliding window fusion method

Start	Range	Operation	融合方法
1	1	—	Min rule
N	1	—	Max rule
—	N	Sum	Average rule
—	N	Product	Product rule

图3 本文所提医学分类算法基本流程图

Fig.3 Basic flow chart of medical classification algorithm proposed in this paper

表2 不同方法分类精度对比

Table 2 Comparison of classification accuracy of different methods

分类方法	分类准确率/%
文献[48]	95.4
CNN	96.5
文献[49]	96.7
文献[50]	97.1
本文方法	98.2

图4 乳腺肿块特征可视化分布

Fig.4 Visualized distribution of breast mass features

表3 每次实验样本情况

Table 3 Sample situation for each experiment

频次	训练样本	测试样本
1	150 380	76 502
2	160 920	76 530
3	165 400	79 980
4	177 000	80 640
5	150 820	71 600
6	135 340	68 850
7	144 828	68 877
8	148 860	71 980
9	159 300	72 576
10	135 738	64 440
11	181 940	87 978
12	194 700	88 704

表4 脑肿瘤组织分类结果对比

Table 4 Comparison of classification results of

样本编号	SVM	CNN	文献[51]	本文方法
1	89.41	91.09	92.82	94.70
2	76.71	85.26	89.76	93.48
3	93.32	94.32	95.15	97.12
4	95.29	96.01	97.89	98.88
5	94.31	95.32	97.18	99.16
6	94.87	96.02	97.90	98.89
7	91.05	92.16	93.92	95.82
8	89.87	91.09	93.82	95.70
9	86.54	89.05	91.72	93.55
10	89.33	92.03	93.79	95.69
11	78.23	86.05	90.57	93.30
12	79.31	85.06	89.55	92.26
平均值	88.19	91.12	93.67	95.71

图5 TCIA-CT数据库部分示例图片

Fig.5 Some sample pictures of TCIA-CT database

图6 OASIS-MRI数据库部分示例图片

Fig.6 Some sample pictures of OASIS-MRI database

表5 不同算法在医学图像数据库的分类精度

Table 5 Classification accuracy of different algorithms in medical image database %

方法类型	TCIA-CT	OASIS-MRI
LBP+SVM	71.8	57.5
HOG+KNN	85.1	67.6
HOG+SVM	87.3	81.6
DeepNet1	98.7	89.2
DeepNet3	99.2	92.1
本文方法	100.0	95.9

参考文献 54

[1]	SONG Y, CAI W, HUANG H, et al. Large margin local estimate with applications to medical image classification[J]. IEEE Transactions on Medical Imaging, 2015, 34(6): 1362-1377.
[2]	MOHSEN H, EL-DAHSHAN E S A, EL-HORBATY E S M, et al. Classification using deep learning neural networks for brain tumors[J]. Future Computing and Informatics Journal, 2018, 3(1): 68-71.
[3]	KEVIN Z, GREENSPAN H, SHEN D G, et al. Deep learning for medical image analysis[M]. New York: Academic Press, 2017.
[4]	VERMA J, NATH M, TRIPATHI P, et al. Analysis and identification of kidney stone using Kth nearest neighbour (KNN) and support vector machine (SVM) classification techniques[J]. Pattern Recognition and Image Analysis, 2017, 27(3): 574-580.
[5]	SPANHOL F A, OLIVEIRA L S, CAVALIN P R, et al. Deep features for breast cancer histopathological image classification[C]// Proceedings of the 2017 IEEE International Conference on Systems, Man, and Cybernetics, Banff, Oct 5-8, 2017. Piscataway: IEEE, 2017: 1868-1873.
[6]	SPANHOL F A, OLIVEIRA L S, PETITJEAN C, et al. A dataset for breast cancer histopathological image classification[J]. IEEE Transactions on Biomedical Engineering, 2016, 63(7): 1455-1462.
[7]	ESTEVA A, KUPREL B, NOVOA R A, et al. Dermatologist-level classification of skin cancer with deep neural networks[J]. Nature, 2017, 542(7639): 115-118.
[8]	LIU D, WANG S, HUANG D, et al. Medical image classification using spatial adjacent histogram based on adaptive local binary patterns[J]. Computers in Biology and Medicine, 2016, 72: 185-200.
[9]	MOHAN G, SUBASHINI M M. MRI based medical image analysis: survey on brain tumor grade classification[J]. Biomedical Signal Processing and Control, 2018, 39: 139-161.
[10]	WU J J, YU Y N, HUANG C, et al. Deep multiple instance learning for image classification and auto-annotation[C]// Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition, Boston, Jun 7-12, 2015. Washington: IEEE Computer Society, 2015: 3460-3469.
[11]	RAMTEKE R J, MONALI Y K. Automatic medical image classification and abnormality detection using k-nearest neighbour[J]. International Journal of Advanced Computer Research, 2012, 2(4): 190-196.
[12]	CORSO J J, SHARON E, DUBE S, et al. Efficient multilevel brain tumor segmentation with integrated Bayesian model classification[J]. IEEE Transactions on Medical Imaging, 2008, 27(5): 629-640.
[13]	RAO A L, LEE Y, GASS A, et al. Classification of Alzheimer’s disease from structural MRI using sparse logistic regression with optional spatial regularization[C]// Proceedings of the 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Boston, Aug 30-Sep 3, 2011. Piscataway:IEEE, 2011: 4499-4502.
[14]	NIWAS S I, PALANISAMY P, SUJATHAN K. Complex wavelet as nucleus descriptors for automated cancer cytology classifier system using ANN[C]// Proceedings of the 2010 IEEE International Conference on Computational Intelligence and Computing Research, Coimbatore, Dec 28-29, 2010. Piscataway: IEEE, 2010: 1-5.
[15]	CAMLICA Z, TIZHOOSH H R, KHALVATI F. Medical image classification via SVM using LBP features from saliency-based folded data[C]// Proceedings of the 14th IEEE International Conference on Machine Learning and Applications, Miami, Dec 9-11, 2015. Piscataway: IEEE, 2015: 128-132.
[16]	CSURKA G, DANCE C, FAN L, et al. Visual categorization with bags of keypoints[C]// Proceedings of the 2004 Workshop on Statistical Learning in Computer Vision, Prague, May 16, 2004. Washington:IEEE Computer Society, 2004: 1-2.
[17]	SINGH A. Detection of brain tumor in MRI images, using combination of fuzzy C-means and SVM[C]// Proceedings of the 2nd International Conference on Signal Processing and Integrated Networks, Noida, Feb 19-20, 2015. Piscataway: IEEE, 2015: 98-102.
[18]	ZHANG Y D, CHEN S, WANG S H, et al. Magnetic resonance brain image classification based on weighted type fractional Fourier transform and nonparallel support vector machine[J]. International Journal of Imaging Systems and Technology, 2015, 25(4): 317-327.
[19]	ZHU S, TIZHOOSH H R. Radon features and barcodes for medical image retrieval via SVM[C]// Proceedings of the 2016 International Joint Conference on Neural Networks, Vancouver, Jul 24-29, 2016. Piscataway: IEEE, 2016: 5065-5071.
[20]	ABBASS H A. An evolutionary artificial neural networks approach for breast cancer diagnosis[J]. Artificial Intelligence in Medicine, 2002, 25(3): 265-281.
[21]	KARABATAK M, INCE M C. An expert system for detection of breast cancer based on association rules and neural network[J]. Expert Systems with Applications, 2009, 36(2): 3465-3469.
[22]	JUANG L H, WU M N. MRI brain lesion image detection based on color-converted K-means clustering segmentation[J]. Measurement, 2010, 43(7): 941-949.
[23]	ZHANG D Q, CHEN S C. A novel kernelized fuzzy C-means algorithm with application in medical image segmentation[J]. Artificial Intelligence in Medicine, 2004, 32(1): 37-50.
[24]	SINGH D, KAUR K. Classification of abnormalities in brain MRI images using GLCM, PCA and SVM[J]. International Journal of Engineering and Advanced Technology, 2012, 1(6): 243-248.
[25]	HINTON G E, SALAKHUTDINOV R R. Reducing the dimensionality of data with neural networks[J]. Science, 2006, 313(5786): 504-507.
[26]	RADENOVIĆ F, TOLIAS G, CHUM O. CNN image retrieval learns from BoW: unsupervised fine-tuning with hard examples[C]// LNCS 9905: Proceedings of the 14th European Conference on Computer Vision, Amsterdam, Oct 11-14, 2016. Cham: Springer, 2016: 3-20.
[27]	CAMPOS V, JOU B, GIRÓ-I-NIETO X. From pixels to sentiment: fine-tuning CNNs for visual sentiment prediction[J]. Image and Vision Computing, 2017, 65: 15-22.
[28]	GIRSHICK R B. Fast R-CNN[C]// Proceedings of the 2015 IEEE International Conference on Computer Vision, Santiago, Dec 13-16, 2015. Washington: IEEE Computer Society, 2015: 1440-1448.
[29]	LIN T Y, ROYCHOWDHURY A, MAJI S. Bilinear CNN models for fine-grained visual recognition[C]// Proceedings of the 2015 IEEE International Conference on Computer Vision, Santiago, Dec 7-13, 2015. Washington: IEEE Computer Society, 2015: 1449-1457.
[30]	BROSCH T, TAM R. Alzheimer’s disease neuroimaging initiative Manifold learning of brain MRIs by deep learning[C]// LNCS 8150: Proceedings of the 2013 International Conference on Medical Image Computing and Computer-Assisted Intervention, Nagoya, Sep 22-26, 2013. Berlin, Heidelberg: Springer, 2013: 633-640.
[31]	PLIS S M, HJELM D R, SALAKHUTDINOV R, et al. Deep learning for neuroimaging: a validation study[J]. Frontiers in Neuroscience, 2014, 8: 229-238.
[32]	CHENG J Z, NI D, CHOU Y H, et al. Computer-aided diagnosis with deep learning architecture: applications to breast lesions in US images and pulmonary nodules in CT scans[J]. Scientific Reports, 2016, 6: 24454-24466.
[33]	KALLENBERG M, PETERSEN K, NIELSEN M, et al. Unsupervised deep learning applied to breast density segmentation and mammographic risk scoring[J]. IEEE Transactions on Medical Imaging, 2016, 35(5): 1322-1331.
[34]	TAJBAKHSH N, SHIN J Y, GURUDU S R, et al. Convolutional neural networks for medical image analysis: full training or fine tuning?[J]. IEEE Transactions on Medical Imaging, 2016, 35(5): 1299-1312.
[35]	LIANG Z H, POWELL A, ERSOY I, et al. CNN-based image analysis for malaria diagnosis[C]// Proceedings of the 2016 IEEE International Conference on Bioinformatics and Biomedicine, Shenzhen, Dec 15-18, 2016. Washington: IEEE Computer Society, 2016: 493-496.
[36]	GAO M C, BAGCI U, LU L, et al. Holistic classification of CT attenuation patterns for interstitial lung diseases via deep convolutional neural networks[J]. Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization, 2018, 6(1): 1-6.
[37]	SHIE C K, CHUANG C H, CHOU C N, et al. Transfer representation learning for medical image analysis[C]// Proceedings of the 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Milan, Aug 25-29, 2015. Piscataway: IEEE, 2015: 711-714.
[38]	ZHU Q K, DU B, TURKBEY B, et al. Deeply-supervised CNN for prostate segmentation[C]// Proceedings of the 2017 International Joint Conference on Neural Networks, Anchorage, May 14-19, 2017. Piscataway: IEEE, 2017: 178-184.
[39]	SHEN W, ZHOU M, YANG F, et al. Multi-scale convolutional neural networks for lung nodule classification[C]// LNCS 9123: Proceedings of the 24th International Conference on Information Processing in Medical Imaging, Sabhal Mor Ostaig, Jun 28-Jul 3, 2015. Cham:Springer, 2015: 588-599.
[40]	KAWAHARA J, HAMARNEH G. Multi-resolution-tract CNN with hybrid pretrained and skin-lesion trained layers[C]// LNCS 10019: Proceedings of the 7th International Workshop on Machine Learning in Medical Imaging, Athens, Oct 17, 2016. Cham:Springer, 2016: 164-171.
[41]	BIDART R, GANGEH M J, PEIKARI M, et al. Localization and classification of cell nuclei in post-neoadjuvant breast cancer surgical specimen using fully convolutional networks[C]// Proceedings of the Medical Imaging:Digital Pathology 2018, Houston, Feb 10-15, 2018. San Francisco: SPIE, 2018: 1-8.
[42]	HE K M, ZHANG X Y, REN S Q, et al. Delving deep into rectifiers: surpassing human-level performance on ImageNet classification[C]// Proceedings of the 2015 IEEE International Conference on Computer Vision, Santiago, Dec 7-13, 2015. Piscataway: IEEE, 2015: 1026-1034.
[43]	GLOROT X, BENGIO Y. Understanding the difficulty of training deep feedforward neural networks[C]// Proceedings of the 13th International Conference on Artificial Intelligence and Statistics, Sardinia, May 13-15, 2010: 249-256.
[44]	KLAMBAUER G, UNTERTHINER T, MAYR A, et al. Self-normalizing neural networks[C]// Advances in Neural Information Processing Systems 30, Long Beach, Dec 4-9, 2017. Red Hook: Curran Associates, 2017: 971-980.
[45]	JADERBERG M, SIMONYAN K, ZISSERMAN A. Spatial transformer networks[C]// Advances in Neural Information Processing Systems 28, Montreal, Dec 7-12, 2015. Red Hook: Curran Associates, 2015: 2017-2025.
[46]	WANG X L, GIRSHICK R B, GUPTA A, et al. Non-local neural networks[C]// Proceedings of the 2018 IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, Jun 18-22, 2018. Washington: IEEE Computer Society, 2018: 7794-7803.
[47]	GÜLÇEHRE Ç, MOCZULSKI M, DENIL M, et al. Noisy activation functions[C]// Proceedings of the 33rd International Conference on Machine Learning, New York, Jun 19-24, 2016: 3059-3068.
[48]	XIE W, LI Y, MA Y. Breast mass classification in digital mammography based on extreme learning machine[J]. Neurocomputing, 2016, 173: 930-941.
[49]	ZHENG Y, FAN J, ZHANG J, et al. Hierarchical learning of multi-task sparse metrics for large-scale image classification[J]. Pattern Recognition, 2017, 67: 97-109.
[50]	QIU Y, YAN S, GUNDREDDY R R, et al. A new approach to develop computer-aided diagnosis scheme of breast mass classification using deep learning technology[J]. Journal of X-ray Science and Technology, 2017, 25(5): 751-763.
[51]	MOHSEN H, EL-DAHSHAN E S A, EL-HORBATY E S M, et al. Classification using deep learning neural networks for brain tumors[J]. Future Computing and Informatics Journal, 2018, 3(1): 68-71.
[52]	CLARK K, VENDT B, SMITH K, et al. The cancer imaging archive (TCIA): maintaining and operating a public information repository[J]. Journal of Digital Imaging, 2013, 26(6): 1045-1057.
[53]	MARCUS D S, WANG T H, PARKER J, et al. Open access series of imaging studies (OASIS): cross-sectional MRI data in young, middle aged, nondemented, and demented older adults[J]. Journal of Cognitive Neuroscience, 2007, 19(9): 1498-1507.
[54]	DUBEY S R, SINGH S K, SINGH R K. Local wavelet pattern: a new feature descriptor for image retrieval in medical CT databases[J]. IEEE Transactions on Image Processing, 2015, 24(12): 5892-5903.

权重初始化-滑动窗口CNN的医学图像分类

Medical Image Classification Algorithm Based on Weight Initialization-Sliding Window CNN

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 11

参考文献 54

相关文章 15

编辑推荐

Metrics

[1]	吕晓琦, 纪科, 陈贞翔, 孙润元, 马坤, 邬俊, 李浥东. 结合注意力与循环神经网络的专家推荐算法[J]. 计算机科学与探索, 2022, 16(9): 2068-2077.
[2]	张祥平, 刘建勋. 基于深度学习的代码表征及其应用综述[J]. 计算机科学与探索, 2022, 16(9): 2011-2029.
[3]	李冬梅, 罗斯斯, 张小平, 许福. 命名实体识别方法研究综述[J]. 计算机科学与探索, 2022, 16(9): 1954-1968.
[4]	任宁, 付岩, 吴艳霞, 梁鹏举, 韩希. 深度学习应用于目标检测中失衡问题研究综述[J]. 计算机科学与探索, 2022, 16(9): 1933-1953.
[5]	杨才东, 李承阳, 李忠博, 谢永强, 孙方伟, 齐锦. 深度学习的图像超分辨率重建技术综述[J]. 计算机科学与探索, 2022, 16(9): 1990-2010.
[6]	曾凡智, 许露倩, 周燕, 周月霞, 廖俊玮. 面向智慧教育的知识追踪模型研究综述[J]. 计算机科学与探索, 2022, 16(8): 1742-1763.
[7]	刘艺, 李蒙蒙, 郑奇斌, 秦伟, 任小广. 视频目标跟踪算法综述[J]. 计算机科学与探索, 2022, 16(7): 1504-1515.
[8]	赵小明, 杨轶娇, 张石清. 面向深度学习的多模态情感识别研究进展[J]. 计算机科学与探索, 2022, 16(7): 1479-1503.
[9]	夏鸿斌, 肖奕飞, 刘渊. 融合自注意力机制的长文本生成对抗网络模型[J]. 计算机科学与探索, 2022, 16(7): 1603-1610.
[10]	孙方伟, 李承阳, 谢永强, 李忠博, 杨才东, 齐锦. 深度学习应用于遮挡目标检测算法综述[J]. 计算机科学与探索, 2022, 16(6): 1243-1259.
[11]	刘雅芬, 郑艺峰, 江铃燚, 李国和, 张文杰. 深度半监督学习中伪标签方法综述[J]. 计算机科学与探索, 2022, 16(6): 1279-1290.
[12]	程卫月, 张雪琴, 林克正, 李骜. 融合全局与局部特征的深度卷积神经网络算法[J]. 计算机科学与探索, 2022, 16(5): 1146-1154.
[13]	童敢, 黄立波. Winograd快速卷积相关研究综述[J]. 计算机科学与探索, 2022, 16(5): 959-971.
[14]	钟梦圆, 姜麟. 超分辨率图像重建算法综述[J]. 计算机科学与探索, 2022, 16(5): 972-990.
[15]	裴利沈, 赵雪专. 群体行为识别深度学习方法研究综述[J]. 计算机科学与探索, 2022, 16(4): 775-790.