[1]吴越,钱善华,邓大建,等.基于X光吞咽视频的舌运动轨迹仿真研究[J].中国医学物理学杂志,2020,37(9):1130-1137.[doi:10.3969/j.issn.1005-202X.2020.09.010]
 WU Yue,QIAN Shanhua,DENG Dajian,et al.Tongue motion trajectory simulation based on the X-ray video of swallowing angiography[J].Chinese Journal of Medical Physics,2020,37(9):1130-1137.[doi:10.3969/j.issn.1005-202X.2020.09.010]
点击复制

基于X光吞咽视频的舌运动轨迹仿真研究()
分享到:

《中国医学物理学杂志》[ISSN:1005-202X/CN:44-1351/R]

卷:
37
期数:
2020年第9期
页码:
1130-1137
栏目:
医学人工智能
出版日期:
2020-09-25

文章信息/Info

Title:
Tongue motion trajectory simulation based on the X-ray video of swallowing angiography
文章编号:
1005-202X(2020)09-1130-08
作者:
吴越1钱善华1邓大建2王志勇3
1. 江南大学机械工程学院,江苏无锡214122;2. 驻重庆地区第五军代室,重庆巴南400054;3. 福建医科大学第一附属医院,福建 福州350005
Author(s):
WU Yue1 QIAN Shanhua1 DENG Dajian2 WANG Zhiyong3
1. School of Mechanical Engineering, Jiangnan University, Wuxi 214122, China 2. The Fifth Military Commission Office in Chongqing District, Banan 400054, China 3. The First Affiliated Hospital of Fujian Medical University, Fuzhou 350005, China
关键词:
食物吞咽舌根点轨迹方程神经网络
Keywords:
food swallowing tongue root point trajectory equation neural network
分类号:
R318;Q811
DOI:
10.3969/j.issn.1005-202X.2020.09.010
文献标志码:
A
摘要:
为提高吞咽障碍患者的食物吞咽质量,须深入研究口腔环境下食物吞咽的舌运动规律;基于口腔吞咽的X-光运动 视频,通过提取特征图像开展舌吞咽运动的规律研究,采用3种不同的方法开展舌运动轨迹方程的仿真研究。结果表明, 舌运动是由舌根点和运动点组成,即舌根点轨迹类似于椭圆,而舌运动点的轨迹类似于往复运动的直线;多元非线性曲线 快速拟合法、Lsqnonlin 优化5 次多项式函数和BP神经网络法3 种方法的拟合方程的轨迹跟原始轨迹的相关系数分别为 0.85、0.87和0.91,发现神经网络模型更适用于天然舌运动轨迹的研究,这将为功能食品研发和吞咽仿生装置设计提供技 术参考。
Abstract:
To improve the quality of swallowing for the patients with dysphagia, it is necessary to analyze the tongue motions during swallowing. Based on the X-ray video of swallowing angiography, the characteristic images are extracted for analyzing the tongue motions during swallowing. Herein 3 different methods are used to investigate the tongue motion trajectory equation. The results indicated that the tongue motions are composed of tongue root point and several motor points. The motion of tongue root point is approximate to an ellipse, while those of tongue motor points are similar to reciprocating straight lines. The correlation coefficients between original trajectory and the simulated trajectories obtained by the fitting equations of 3 methods, namely multiple nonlinear curve-fitting method, Lsqnonlin optimization of 5 level polynomial function and back-propagation neural network method, were 0.85, 0.87 and 0.91, respectively. The neural network model is found to be more suitable for the simulation of tongue motion trajectory. These results will provide valuable technical reference for the development of functional food and swallowing device in vitro.

备注/Memo

备注/Memo:
【收稿日期】2020-03-26 【基金项目】国家自然科学基金(51775244);江苏省自然科学基金 (BK20161134);中国博士后科学基金(2018T110439);江 苏省研究生科研与实践创新计划(SJCX19_0785) 【作者简介】吴越,研究生,研究方向:仿生结构设计,E-mail: wuyue130622 @126.com 【通信作者】钱善华,博士(后),副教授,研究方向:仿生机械设计,Email: qianjnwx@126.comcn
更新日期/Last Update: 2020-09-25