JZUS - Journal of Zhejiang University SCIENCE

Frontiers of Information Technology & Electronic Engineering 2020 Vol.21 No.8 P.1161-1170

http://doi.org/10.1631/FITEE.1900210

Texture branch network for chronic kidney disease screening based on ultrasound images

Author(s): Peng-yi Hao, Zhen-yu Xu, Shu-yuan Tian, Fu-li Wu, Wei Chen, Jian Wu, Xiao-nan Luo
Affiliation(s): College of Computer Science and Technology, Zhejiang University of Technology, Hangzhou 310023, China; more
Corresponding email(s): fuliwu@zjut.edu.cn
Key Words: Chronic kidney disease, Ultrasound, Texture branch network, Transfer learning

Share this article to： More <<< Previous Article \|Next Article >>>

Abstract: chronic kidney disease (CKD) is a widespread renal disease throughout the world. Once it develops to the advanced stage, serious complications and high risk of death will follow. Hence, early screening is crucial for the treatment of CKD. Since ultrasonography has no side effects and enables radiologists to dynamically observe the morphology and pathological features of the kidney, it is commonly used for kidney examination. In this study, we propose a novel convolutional neural network (CNN) framework named the texture branch network to screen CKD based on ultrasound images. This introduces a texture branch into a typical CNN to extract and optimize texture features. The model can automatically generate texture features and deep features from input images, and use the fused information as the basis of classification. Furthermore, we train the base part of the network by means of transfer learning, and conduct experiments on a dataset with 226 ultrasound images. Experimental results demonstrate the effectiveness of the proposed approach, achieving an accuracy of 96.01% and a sensitivity of 99.44%.

纹理分支网络：基于超声影像的慢性肾脏病筛查模型

郝鹏翼^1,5，徐震宇¹，田树元²，吴福理^1,5，陈为^3,5，吴健^4,5，罗笑南⁶
¹浙江工业大学计算机科学与技术学院，中国杭州市，310023
²浙江省立同德医院，中国杭州市，310012
³浙江大学附属第一医院，中国杭州市，310003
⁴浙江大学计算机科学与技术学院，中国杭州市，310027
⁵浙江大学睿医人工智能研究中心，中国杭州市，310027
⁶桂林电子科技大学人工智能学院，中国桂林市，541004

摘要：慢性肾脏病是一种在世界范围内广泛存在的肾脏疾病。该疾病一旦发展到晚期，伴随而来的是严重并发症与较高死亡风险。因此，早期筛查对于慢性肾脏病诊治至关重要。超声作为一种无创方法，能动态观察肾脏形态和病理特征，常用于肾脏检查。本文提出一种新的卷积神经网络模型，称为纹理分支网络，基于超声影像作慢性肾脏病筛查。该模型通过在经典卷积神经网络中引入纹理分支来提取和优化纹理特征，可自动生成输入图像的纹理特征和深度特征，并使用融合信息进行分类。此外，通过迁移学习训练网络的主干部分，并在具有226张超声影像的数据集上开展实验。实验结果表明，该模型准确率和敏感度分别达到96.01%和99.44%，在慢性肾脏病筛查上具有一定有效性。

关键词：慢性肾脏病；超声；纹理分支网络；迁移学习

Reference

[1]Acharya UR, Meiburger KM, Koh JEW, et al., 2019. Automated detection of chronic kidney disease using higher-order features and elongated quinary patterns from B-mode ultrasound images. Neur Comput Appl, in press.

[2]Ahmad M, Tundjungsari V, Widianti D, et al., 2017. Diagnostic decision support system of chronic kidney disease using support vector machine. Proc 2^nd Int Conf on Informatics and Computing, p.1-4.

[3]Batra A, Batra U, Singh V, 2016. A review to predictive methodology to diagnose chronic kidney disease. Proc 3^rd Int Conf on Computing for Sustainable Global Development, p.2760-2763.

[4]Chang EH, Chong WK, Kasoji SK, et al., 2017. Diagnostic accuracy of contrast-enhanced ultrasound for characterization of kidney lesions in patients with and without chronic kidney disease. BMC Nephrol, 18, Article 266.

[5]Charleonnan A, Fufaung T, Niyomwong T, et al., 2016. Predictive analytics for chronic kidney disease using machine learning techniques. Proc Management and Innovation Technology Int Conf, p.MIT-80-MIT-83.

[6]Chen H, Dou Q, Ni D, et al., 2015. Automatic fetal ultrasound standard plane detection using knowledge transferred recurrent neural networks. Proc 18^th Int Conf on Medical Image Computing and Computer-Assisted Intervention, p.507-514.

[7]Dalal N, Triggs B, 2005. Histograms of oriented gradients for human detection. Proc IEEE Computer Society Conf on Computer Vision and Pattern Recognition, p.886-893.

[8]Dhindsa K, Smail LC, McGrath M, et al., 2018. Grading prenatal hydronephrosis from ultrasound imaging using deep convolutional neural networks. Proc 15^th Conf on Computer and Robot Vision, p.80-87.

[9]Ecder T, 2013. Early diagnosis saves lives: focus on patients with chronic kidney disease. Kidney Int Suppl, 3(4):335-336.

[10]El Nahas AM, Bello AK, 2005. Chronic kidney disease: the global challenge. Lancet, 365(9456):331-340.

[11]Haralick RM, Shanmugam K, Dinstein I, 1973. Textural features for image classification. IEEE Trans Syst Man Cybern, SMC-3(6):610-621.

[12]He KM, Zhang XY, Ren SQ, et al., 2015. Delving deep into rectifiers: surpassing human-level performance on ImageNet classification. Proc IEEE Int Conf on Computer Vision, p.1026-1034.

[13]He KM, Zhang XY, Ren SQ, et al., 2016. Deep residual learning for image recognition. Proc IEEE Conf on Computer Vision and Pattern Recognition, p.770-778.

[14]Ho CY, Pai TW, Peng YC, et al., 2012. Ultrasonography image analysis for detection and classification of chronic kidney disease. Proc 6^th Int Conf on Complex, Intelligent, and Software Intensive Systems, p.624-629.

[15]Hsieh JW, Lee CH, Chen YC, et al., 2014. Stage classification in chronic kidney disease by ultrasound image. Proc 29^th Int Conf on Image and Vision Computing New Zealand, p.271-276.

[16]Iqbal F, Pallewatte AS, Wansapura JP, 2017. Texture analysis of ultrasound images of chronic kidney disease. Proc 17^th Int Conf on Advances in ICT for Emerging Regions, p.1-5.

[17]Jeewantha RA, Halgamuge MN, Mohammad A, et al., 2017. Classification performance analysis in medical science: using kidney disease data. Proc Int Conf on Big Data Research, p.1-6.

[18]Kunwar V, Chandel K, Sabitha AS, et al., 2016. Chronic kidney disease analysis using data mining classification techniques. Proc 6^th Int Conf on Cloud System and Big Data Engineering, p.300-305.

[19]Levey AS, Eckardt KU, Tsukamoto Y, et al., 2005. Definition and classification of chronic kidney disease: a position statement from kidney disease: improving global outcomes (KDIGO). Kidney Int, 67(6):2089-2100.

[20]López V, Fernández A, García S, et al., 2013. An insight into classification with imbalanced data: empirical results and current trends on using data intrinsic characteristics. Inform Sci, 250:113-141.

[21]Oquab M, Bottou L, Laptev I, et al., 2014. Learning and transferring mid-level image representations using convolutional neural networks. Proc IEEE Conf on Computer Vision and Pattern Recognition, p.1717-1724.

[22]Pujari RM, Hajare VD, 2014. Analysis of ultrasound images for identification of chronic kidney disease stages. Proc 1^st Int Conf on Networks & Soft Computing, p.380-383.

[23]Sharma K, Virmani J, 2016. Classification of renal diseases using first order and higher order statistics. Proc 3^rd Int Conf on Computing for Sustainable Global Development, p.425-430.

[24]Shen W, Zhou M, Yang F, et al., 2015. Multi-scale convolutional neural networks for lung nodule classification. Proc 24^th Int Conf on Information Processing in Medical Imaging, p.588-599.

[25]Shin HC, Roth HR, Gao MC, et al., 2016. Deep convolutional neural networks for computer-aided detection: CNN architectures, dataset characteristics and transfer learning. IEEE Trans Med Imag, 35(5):1285-1298.

[26]Subramanya MB, Kumar V, Mukherjee S, et al., 2015. SVM-based CAC system for B-mode kidney ultrasound images. J Dig Imag, 28(4):448-458.

[27]Telea A, 2004. An image inpainting technique based on the fast marching method. J Dig Imag, 9(1):23-34.

[28]Zheng Q, Furth SL, Tasian GE, et al., 2019. Computer-aided diagnosis of congenital abnormalities of the kidney and urinary tract in children based on ultrasound imaging data by integrating texture image features and deep transfer learning image features. J Pediatr Urol, 15(1):75.e1-75.e7.

Open peer comments: Debate/Discuss/Question/Opinion

<1>

Similar articles

- Go to

纹理分支网络：基于超声影像的慢性肾脏病筛查模型

Darkslateblue:Affiliate; Royal Blue:Author; Turquoise:Article

Reference