发布时间: 2018-10-16 摘要点击次数: 全文下载次数: DOI: 10.11834/jig.170683 2018 | Volume 23 | Number 10 医学图像处理

 收稿日期: 2017-12-30; 修回日期: 2018-04-23 第一作者简介: 李琼, 1993年生, 女, 硕士研究生, 主要研究方向为基于深度学习的医疗图像处理。E-mail:1668385998@qq.com;刘莹芳, 女, 硕士研究生, 主要研究方向为基于深度学习的医疗图像处理。E-mail:2449693058@qq.com. 中图法分类号: TP391.4 文献标识码: A 文章编号: 1006-8961(2018)10-1594-10

# 关键词

Automated classification of diabetic retinal images by using deep learning method
Li Qiong, Bai Zhengyao, Liu Yingfang
School of Information Science and Engineering, Yunnan University, Kunming 650500, China

# Abstract

Objective Diabetic retinopathy(DR) is a serious eye disease that causes blindness.The retinal pathological image is an important criterion for diagnosing eye diseases, and the accurate classification of retinal images is a crucial step taken by doctors in developing personalized treatment plans.The automated classification of diabetic retinopathy images has significant clinical values.The traditional image classification methods based on manually extracted features have problems, including complex retinal image processing, discriminative features extraction difficulties, poor classification performance, time-consuming, and difficult objective and consistent diagnoses.In this paper, an improved deep convolutional neural network based on AlexNet and a deep classifier are proposed to realize automated diabetic retinopathy image classification. Method First, training the retinal samples is insufficient because retinal images contain much noise, and the differences between retinal pathological images at adjacent stages are small.These prevent the application of convolutional neural networks in retinal images classification, and retinal images should be preprocessed before they are used as training samples.Preprocessing mainly includes retinal image denoising, enhancement, and normalization.The small number of retinal images and data imbalance at different stages are solved by data enhancement.Second, the feature extraction network is designed based on the network structure of AlexNet.The data distribution is changed during the training process.Hence, a batch normalization layer is introduced before every convolutional layer and fully connected layer of the AlexNet network to produce a new deep network, which we call the BNnet network.The introduction of a batch normalization layer can accelerate the convergence of the network, improve the classification accuracy of the obtained model, and reduce the need for a dropout layer.The BNnet network is a complex deep convolutional neural network, which not only serves as a feature extraction network for retinal images but also effectively suppresses data distribution changes in the training process.In this work, the BNnet network is pre-trained using the ILSVRC2012 dataset based on the transfer learning strategy, and the obtained model is migrated to the enhanced diabetic retinopathy dataset for further study to capture the distinguishing features.Finally, a classifier is designed based on the fully connected layer, which can map the learned deep features to the sample label space.The classifier is composed of the fully connected, ReLU, and dropout layers and is applied in learning to partition a diabetic retinopathy status to no DR, mild DR, moderate DR, severe DR, and proliferative DR. Result We designed four groups of comparative experiments to fully describe the effects of the different depths of neural network and the different training methods, the introduced batch normalized layer, and data preprocessing of the experimental results.The experimental results show that the more layers the network has, the more features are learned with sufficient training samples, and the classification performance of the pre-trained network is better than the traditional direct training method.Moreover, the proposed BNnet neural network and training method can capture the differences of various stages of diabetic retinopathy with a classification accuracy of up to 93%, outperforming other methods.The introduction of the batch normalization layer can control the data distribution changes during the training process and improve the recognition rate.In the case of insufficient retinal image samples, the adoption of transfer learning and data enhancement strategies is good for extracting deep discriminative features for classification.Hence, a deep classifier that is composed of fully connected layers can accurately distinguish the stage when a retinal image is located, indicating that the obtained deep features based on BNnet and transfer learning can provide suitable information for classifiers to accurately classify retinal images into five categories. Conclusion We use deep learning methods to achieve the automatic classification of retinal images.We also present a new diabetic retinopathy classification framework that mainly benefits from three important components:the image preprocessing stage, the deep features extraction stage that is based on transfer learning strategies and BNnet neural networks, and the stable classification stage.Intensive dropout and ReLU are used to suppress the over-fitting problem of the deep learning algorithm when the training samples are insufficient.Experimental results show that deep features combined with the proposed methods can provide suitable information for the construction of the most accurate prediction mode, predict diabetic retinopathy status, and effectively avoid the limitations of manual feature extraction and image classification.This method has relatively better robustness and generalization, and it can be widely used for various image classification problems.In future studies, we will likely develop a real-time computer-aided diagnosis system for diabetic retina images based on the above mentioned approach.

# Key words

diabetic retinopathy image classification; convolution neural network; deep learning; transfer learning; deep features

# 0 引言

1) 图像的质量。采集的视网膜图像的质量容易受到光照、镜头、机器设备和图像采集人员的经验等其他许多不可控因素的影响；

2) 医生的个人经验。医生通常通过视觉检查视网膜图像来评估判定视网膜的病变程度，但是在视网膜病变图像中人眼识别出来的特征是有限的，而且由于医生的临床经验不同，对于同一幅视网膜图像，不同的医生可能给出不同的临床诊断结果；

3) 病理图像自身的特点。视网膜图像不同阶段之间的差异性很小，给特征提取和分类工作带来了一定困难。

# 1.1 卷积神经网络

 $A_j^l = f\left( {\sum\limits_{i = 1}^{M\left( {l - 1} \right)} {A_i^{l - 1}*k_{ij}^l + b_j^l} } \right)$ (1)

# 1.3 批归一化(BN)

 ${{\hat x}^{\left( k \right)}} = \frac{{{x^{\left( k \right)}} - E\left[ {{x^{\left( k \right)}}} \right]}}{{\sqrt {V\left[ {{x^{\left( k \right)}}} \right]} }}$ (2)

 ${y^{\left( k \right)}} = {\gamma ^{\left( k \right)}}{{\hat x}^{\left( k \right)}} + {\beta ^{\left( k \right)}}$ (3)

# 4.2 评价指标

 $R = \frac{{{N_{{\rm{rec}}}}}}{{{N_{{\rm{total}}}}}}$ (5)

# 4.3 实验结果分析

Table 1 Comparison of recognition rate of different training methods

 网络结构和训练方法 $R$ 文献[1] CompactNet 0.69 本文方法 LeNet+增强数据 0.43 AlexNet+增强数据 0.63 BNnet+增强数据 0.71 BNnet+迁移学习+原始数据 0.65 BNnet+迁移学习+增强数据 0.90 BNnet+迁移学习+增强数据+分类器 0.93

# 参考文献

• [1] Ding P L, Li Q Y, Zhang Z, et al. Diabetic retinal image classification method based on deep neural network[J]. Journal of Computer Applications, 2017, 37(3): 699–704. [丁蓬莉, 李清勇, 张振, 等. 糖尿病性视网膜图像的深度神经网络分类方法[J]. 计算机应用, 2017, 37(3): 699–704. ] [DOI:10.11772/j.issn.1001-9081.2017.03.699]
• [2] He X Y, Han Z Y, Wei B Z. Breast cancer histopathological image auto-classification using deep learning[J]. Computer Engineering and Applications, 2017. [何雪英, 韩忠义, 魏本征. 基于深度学习的乳腺癌病理图像自动分类[J]. 计算机工程与应用, 2017. ] [DOI:10.3778/j.issn.1002-8331.1701-0392]
• [3] Wu G R, Kim M, Wang Q, et al. Scalable high-performance image registration framework by unsupervised deep feature representations learning[J]. IEEE Transactions on Biomedical Engineering, 2016, 63(7): 1505–1516. [DOI:10.1109/TBME.2015.2496253]
• [4] Nguyen P, Tran T, Wickramasinghe N, et al. Deepr:A convolutional net for medical records[J]. IEEE Journal of Biomedical and Health Informatics, 2017, 21(1): 22–30. [DOI:10.1109/JBHI.2016.2633963]
• [5] Guo Y R, Gao Y Z, Shen D G. Deformable MR prostate segmentation via deep feature learning and sparse patch matching[J]. IEEE Transactions on Medical Imaging, 2016, 35(4): 1077–1089. [DOI:10.1109/TMI.2015.2508280]
• [6] Miotto R, Li L, Kidd B A, et al. Deep patient:an unsupervised representation to predict the future of patients from the electronic health records[J]. Scientific Reports, 2016, 6: #26094. [DOI:10.1038/srep26094]
• [7] Li Q, Cai W D, Wang X G, et al.Medical image classification with convolutional neural network[C]//Proceedings of the 13th International Conference on Control Automation Robotics & Vision.Singapore: IEEE, 2014: 844-848.[DOI: 10.1109/ICARCV.2014.7064414]
• [8] Moeskops P, Viergever M A, Mendrik A M, et al. Automatic segmentation of MR brain images with a convolutional neural network[J]. IEEE Transactions on Medical Imaging, 2016, 35(5): 1252–1261. [DOI:10.1109/TMI.2016.2548501]
• [9] Shen D G, Wu G R, Suk H I. Deep learning in medical image analysis[J]. Annual Review of Biomedical Engineering, 2017, 19: 221–248. [DOI:10.1146/annurev-bioeng-071516-044442]
• [10] Pan S J, Yang Q. A survey on transfer learning[J]. IEEE Transactions on Knowledge and Data Engineering, 2010, 22(10): 1345–1359. [DOI:10.1109/TKDE.2009.191]
• [11] Bengio Y, Courville A, Vincent P. Representation learning:a review and new perspectives[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(8): 1798–1828. [DOI:10.1109/TPAMI.2013.50]
• [12] Ioffe S, Szegedy C.Batch normalization: accelerating deep network training by reducing internal covariate shift[C]//Proceedings of the 32nd International Conference on International Conference on Machine Learning.Lille, France: PMLR, 2015: 448-456.
• [13] Huang C, He Z H, Cao G T, et al. Task-driven progressive part localization for fine-grained object recognition[J]. IEEE Transactions on Multimedia, 2016, 18(12): 2372–2383. [DOI:10.1109/TMM.2016.2602060]
• [14] He K M, Zhang X Y, Ren S Q, et al. Spatial pyramid pooling in deep convolutional networks for visual recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 37(9): 1904–1916. [DOI:10.1109/TPAMI.2015.2389824]
• [15] Srivastava N, Hinton G, Krizhevsky A, et al. Dropout:a simple way to prevent neural networks from overfitting[J]. Journal of Machine Learning Research, 2014, 15(1): 1929–1958.
• [16] Nair V, Hinton G E.Rectified linear units improve restricted Boltzmann machines[C]//Proceedings of the 27th International Conference on International Conference on Machine Learning.Haifa, Israel: Omnipress, 2010: 807-814.
• [17] 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. [DOI:10.1109/TBME.2015.2496264]
• [18] Zeiler M D, Fergus R.Visualizing and understanding convolutional networks[C]//Proceedings of the 13th European Conference on Computer Vision.Zurich, Switzerland: Springer, 2014: 818-833.[DOI: 10.1007/978-3-319-10590-1_53]