一种基于多模态特征融合的骨质疏松评估方法

doi:10.13190/j.jbupt.2019-150

北京邮电大学学报 ›› 2019, Vol. 42 ›› Issue (6): 84-90.doi: 10.13190/j.jbupt.2019-150

一种基于多模态特征融合的骨质疏松评估方法

罗涛, 李剑峰, 韩家辉, 王艺宁, 雷璐

1. 北京邮电大学北京先进信息网络实验室, 北京 100876;
2. 北京邮电大学网络体系构建与融合北京市重点实验室, 北京 100876

收稿日期:2019-07-10 出版日期:2019-12-28 发布日期:2019-11-15
作者简介:罗涛(1971-),男,教授,博士生导师,E-mail:tluo@bupt.edu.cn.
基金资助:
国家重点研发计划重点专项项目（2016YFF0201003）；面向智慧医疗的个人信息保护关键技术及应用（201702017）

Osteoporosis Evaluation Method Based on Multimodal Feature Fusion

LUO Tao, LI Jian-feng, HAN Jia-hui, WANG Yi-ning, LEI Lu

1. Beijing Laboratory of Advanced Information Networks, Beijing University of Posts and Telecommunications, Beijing 100876, China;
2. Beijing Key Laboratory of Network System Architecture and Convergence, Beijing University of Posts and Telecommunications, Beijing 100876, China

Received:2019-07-10 Online:2019-12-28 Published:2019-11-15

摘要/Abstract

摘要： 针对现有骨质疏松评估中诊断依据单一、准确率低的问题，综合考虑骨骼图像数据和问卷数据，首先提出一种基于深度神经网络的多模态特征融合骨质疏松评估方法；然后，针对骨骼图像特征较浅、结构固定的特点，使用Unet进行图像分割预处理，去除冗余信息以提升分类准确性；最后，针对普通卷积操作在把握全局信息方面的不足，提出采用基于non-local模块的卷积神经网络来进一步丰富特征信息.交叉验证结果表明，提出的多模态特征融合方法与仅单独使用图像数据或问卷数据的机器学习方法相比具有明显的优势，分类准确率分别提升了3.2%和22.3%.

关键词: 骨质疏松, 多模态融合, 深度神经网络

Abstract: Aiming at the problems that the problems of single diagnosis and low accuracy in the existing osteoporosis assessment, considering the bone image data and questionnaire data, a multi-modal feature fusion osteoporosis evaluation method based on deep neural network was proposed. And, for the characteristics of shallow image and fixed structure of bone image, Unet is used to perform image segmentation preprocessing to remove redundant information. In view of the shortcomings of ordinary convolution operations in grasping the global information, a new convolutional neural network based on non-local module was proposed to further enrich the feature information. Cross-validation shows that the proposed multimodal feature fusion method has obvious advantages compared with the machine learning method using only image data or questionnaire data alone, and the classification accuracy rate is increased by 3.2% and 22.3%.

Key words: osteoporosis, multi-modal fusion, deep neural network

中图分类号:

TP183

罗涛, 李剑峰, 韩家辉, 王艺宁, 雷璐. 一种基于多模态特征融合的骨质疏松评估方法[J]. 北京邮电大学学报, 2019, 42(6): 84-90.

LUO Tao, LI Jian-feng, HAN Jia-hui, WANG Yi-ning, LEI Lu. Osteoporosis Evaluation Method Based on Multimodal Feature Fusion[J]. JOURNAL OF BEIJING UNIVERSITY OF POSTS AND TELECOM, 2019, 42(6): 84-90.

参考文献

[1] Cruz A S, Lins H C, Medeiros R V A, et al. Artificial intelligence on the identification of risk groups for osteoporosis, a general review[J]. BioMedical Engineering OnLine, 2018(17):12.
[2] 郑博, 蒋涛, 黄佳涌, 等. 我国中老年脆性髋部骨折及桡骨远端骨折骨质疏松诊断标准与WHO诊断标准的比较分析[J]. 中国骨质疏松杂志, 2018, 24(10):1299-1304. Zheng Bo, Jiang Tao, Huang Jiayong, et al. Comparison of osteoporosis diagnostic criteria of China and WHO in Chinese patients with brittle hip fracture or distal radius fracture[J]. Chinese Journal of Osteoporosis, 2018, 24(10):1299-1304.
[3] 吴非同, 胡传来, 方炎福, 等. 社区中老年人群骨质疏松影响因素研究[J]. 中国全科医学, 2011, 14(14):1540-1542. Wu Feitong, Hu Chuanlai, Fang Yanfu, et al. Prevalence and influencing factors of osteoporosis among mid-aged and elderly community residents[J]. Chinese General Practice, 2011, 14(14):1540-1542.
[4] 何俊勇, 何礼霞, 王佑娟, 等. 男性体检人群骨密度调查及其影响因素分析[J]. 实用医院临床杂志, 2018, 15(3):78-80. He Junyong, He Lixia, Wang Youjuan, et al. A survey on bone mineral density and influencing factors in male health examination population[J]. Practical Journal of Clinical Medicine, 2018, 15(3):78-80.
[5] 李小松, 杜文彦. 200例正常老年人性别及相关因素对腰椎骨密度的影响[J]. 中国骨质疏松杂志, 2017, 23(5):616-622. Li Xiaosong, Du Wenyan. Effects of gender and related factors on lumbar spine bone mineral density in 200 cases of healthy elderly people[J]. Chinese Journal of Osteoporosis, 2017, 23(5):616-622.
[6] 翁改志, 路军梅, 孙朝阳, 等. 绝经期女性骨密度调查及相关因素分析[J]. 重庆医科大学学报, 2017, 42(2):153-157. Weng Gaizhi, Lu Junmei, Sun Chaoyang, et al. Investigation on bone mineral density among menopause women and its influencing factors[J]. Journal of Chongqing Medical University, 2017, 42(2):153-157.
[7] Suzuki T, Shimoda T, Takahashi N, et al. Factors affecting bone mineral density among snowy region residents in Japan:analysis using multiple linear regression and Bayesian network model[J]. Interactive Journal of Medical Research, 2018, 7(1):e10.
[8] Kim J W, Ha Yongchan, Lee Y K. Factors affecting bone mineral density measurement after fracture in South Korea[J]. Journal of Bone Metabolism, 2017, 24(4):217.
[9] Wang Xiaolong, Girshick R, Gupta A, et al. Non-local neural networks[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York:IEEE Press, 2018:7794-7803.
[10] Ronneberger O, Fischer P, Brox T. U-net:convolutional networks for biomedical image segmentation[M]//Lecture Notes in Computer Science. Cham:Springer International Publishing, 2015:234-241.
[11] 黄凯. 骨质疏松症社区筛查适宜技术研究[D]. 北京:首都医科大学, 2013.
[12] 方骁然, 应忍冬, 梁伟, 等. 基于SVM的骨质疏松识别算法[J]. 信息技术, 2014, 38(11):39-41, 45. Fang Xiaoran, Ying Rendong, Liang Wei, et al. Classification of osteoporosis by SVM based algorithm[J]. Information Technology, 2014, 38(11):39-41, 45.
[13] Mohamed E I, Meshref R A, Abdel-Mageed S M, et al. A novel morphological analysis of DXA-DICOM images by artificial neural networks for estimating bone mineral density in health and disease[J]. Journal of Clinical Densitometry, 2019, 22(3):382-390.
[14] 余锦娟, 林勇. 基于机器学习的骨质疏松性骨折预测研究[J]. 中国医学物理学杂志, 2018, 35(11):1329-1333. Yu Jinjuan, Lin Yong. Prediction of osteoporotic fractures based on machine learning[J]. Chinese Journal of Medical Physics, 2018, 35(11):1329-1333.

一种基于多模态特征融合的骨质疏松评估方法

Osteoporosis Evaluation Method Based on Multimodal Feature Fusion

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 4

编辑推荐

Metrics

本文评价

[1]	王延凯龙华邵玉斌杜庆治王瑶. 基于融合特征MGCC的语种识别方法[J]. 北京邮电大学学报, 2023, 46(2): 116-121.
[2]	李锦珑陈琼琼丁志杰刘振宇. 基于面部行为与语音融合特征的重度抑郁识别[J]. 北京邮电大学学报, 2023, 46(1): 32-37.
[3]	韩兴, 刘晓平, 王刚, 韩松. 基于深度神经网络复杂场景下的机器人拣选方法[J]. 北京邮电大学学报, 2019, 42(5): 22-28.
[4]	董枫, 刘天铭, 徐国爱, 郭燕慧, 李承泽. 面向Android二进制代码的缺陷预测方法[J]. 北京邮电大学学报, 2018, 41(1): 13-23.