[1]李海岩,张欣玉,可婷,等.基于深度学习的汽车乘员腰椎损伤预测及影响因素分析[J].中国医学物理学杂志,2025,42(3):388-396.[doi:10.3969/j.issn.1005-202X.2025.03.016]
　LI Haiyan,ZHANG Xinyu,et al.Prediction of occupant lumbar spine injuries based on machine learning and analysis of influencing factors[J].Chinese Journal of Medical Physics,2025,42(3):388-396.[doi:10.3969/j.issn.1005-202X.2025.03.016]
点击复制

基于深度学习的汽车乘员腰椎损伤预测及影响因素分析()
分享到：

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

卷:
42
期数:
2025年第3期
页码:
388-396
栏目:
生物材料与力学
出版日期:
2025-03-20

文章信息/Info

Title:
Prediction of occupant lumbar spine injuries based on machine learning and analysis of influencing factors
文章编号:
1005-202X(2025)03-0388-09
作者:
李海岩 1; 2; 张欣玉 1; 2; 可婷 3; 王彦鑫 1; 2; 贺丽娟 1; 2; 吕文乐 1; 2; 崔世海 1; 2; 阮世捷 1; 2
1. 天津科技大学机械工程学院,天津 300222;2. 现代汽车安全技术国际联合研究中心,天津 300222;3. 天津科技大学人工智能 学院,天津 300457
Author(s):
LI Haiyan1;  2;  ZHANG Xinyu1;  2;  KE Ting3;  WANG Yanxin1;  2;  HE Lijuan1;  2;  LU?Wenle1;  2;  CUI Shihai1;  2;  YUAN Shijie1;  2
1. College of Mechanical Engineering, Tianjin University of Science and Technology, Tianjin 300222, China; 2. International Research Association on Emerging Automotive Safety Technology, Tianjin 300222, China; 3. College of Artificial Intelligence, Tianjin University of Science and Technology, Tianjin 300457, China
关键词:
腰椎损伤仿生模型; 损伤机理; 机器学习; 主成分分析; 神经网络
Keywords:
bionic model of lumbar spine injury;  injury mechanism;  machine learning;  principal component analysis;  neural network
分类号:
R318.01
DOI:
10.3969/j.issn.1005-202X.2025.03.016
文献标志码:
A
摘要:
基于 CT 影像数据,构建高生物逼真度的腰椎损伤仿生模型,并对标尸体实验数据验证模型的有效性。解耦汽车正 面碰撞中乘员上躯干由于惯性作用向前俯冲后受约束系统作用回位时与座椅接触所受压迫的历程,设计跌落实验进行仿 真分析。基于深度学习算法对仿真输出结果进行训练预测,并验证训练后所得神经网络预测模型的准确性。采用主成分 分析和交叉逆向方法对关键参数进行相关性分析。结果表明:训练所得腰椎结构损伤预测模型具有较高可靠性(R2> 0.9)。综合分析发现,腰椎结构在受轴向冲击后 L4椎体承受最大冲击载荷,可将其作为腰椎损伤量化评价代表。各环境 变量中,L4腰椎轴向力主要受躯干质量及跌落高度影响,二者均对其具有正相关影响。躯干质量、后倾角度及跌落高度对 内能影响均存在正向影响;而躯干质量及跌落高度对应力影响呈负相关性。该研究结果为进一步理清智能座舱环境中腰 椎损伤机理以制订相应的安全防护策略及汽车乘员安全保护评价等提供科学的参考依据。
Abstract:
Based on CT scan data, a bionic model of lumbar spine injuries with high biofidelity is developed and validated through cadaver experiments. Decoupling the constraint system that affects occupants during collisions due to inertial forces and the subsequent pressure exerted by the seat upon returning to position, a simulated fall experiment is designed. The simulated outcomes are trained and predicted using deep learning algorithms, and the accuracy of the trained neural network prediction model is verified. Key parameters are analyzed for correlation using principal component analysis and crossreverse methods. The results shows that the predicted lumbar spine injury model obtained from training has high reliability (R2>0.9). Comprehensive analysis reveals that after experiencing axial impact, the L4 vertebral body bears the highest impact load and can be used as a representative measure of lumbar spine injury. Among the environmental variables, the axial force on the L4 lumbar spine is mainly affected by torso mass and fall height, both of which have positive correlations. Torso mass, fall height, and posture angle all have positive effects on internal energy. Conversely, torso mass and fall height have negative correlations with stress. These research findings provide a scientific basis for further elucidating lumbar spine injury mechanisms in intelligent cockpit environments, devising corresponding safety protection measures, and evaluating occupant safety in automobiles.

备注/Memo

备注/Memo:
【收稿日期】2024-11-19 【基金项目】国家重点研发计划(2018YFC0807203-1);国家自然科学 基金(81471274,81371360) 【作者简介】李海岩,博士,教授,研究方向:损伤生物力学与汽车安全, E-mail: lihaiyan@tust.edu.cn
更新日期/Last Update: 2025-03-26