首页> 外文会议>2019 Spring Simulation Conference >Towards Musculoskeletal Simulation-Aware Fall Injury Mitigation: Transfer Learning with Deep CNN for Fall Detection

【24h】

Towards Musculoskeletal Simulation-Aware Fall Injury Mitigation: Transfer Learning with Deep CNN for Fall Detection

机译：迈向了解肌肉骨骼模拟的跌倒伤害减轻：采用深度CNN的转移学习进行跌倒检测

获取原文

获取原文并翻译 | 示例

页面导航

摘要
著录项
相似文献
相关主题

摘要

This paper presents early work on a fall detection method using transfer learning method, in conjunction with a long-term effort to combine efficient machine learning and prior personalized musculoskeletal modeling to deploy fall injury mitigation in geriatric subjects. Inspired by the tremendous progress in image-based object recognition with deep convolutional neural networks (DCNNs), we opt for a pre-trained kinematics-based machine learning approach through existing large-scale annotated accelerometry datasets. The accelerometry datasets are converted to images using time-frequency analysis, based on scalograms, by computing the continuous wavelet transform filter bank. Subsequently, data augmentation is performed on these scalogram images to increase accuracy, thereby complementing limited labeled fall sensor data, enabling transfer learning from the existing pre-trained model. The experimental results on publicly available URFD datasets demonstrate that transfer learning leads to a better performance than the existing methods in the case of scarce labeled training data.

机译：本文介绍了使用转移学习方法的跌倒检测方法的早期工作，并结合了长期的努力，以结合有效的机器学习和先前的个性化肌肉骨骼模型，以在老年患者中部署减轻坠落的方法。受基于深度卷积神经网络（DCNN）的基于图像的对象识别的巨大进步的启发，我们选择了一种通过现有的大规模带注释的加速度计数据集进行基于运动学的预先训练的机器学习方法。通过基于连续图，通过计算连续小波变换滤波器组，使用时频分析将加速度计数据集转换为图像。随后，在这些比例尺图像上执行数据增强以提高准确性，从而补充有限的标记跌倒传感器数据，从而能够从现有的预训练模型进行转移学习。在公开可用的URFD数据集上的实验结果表明，在缺乏标签训练数据的情况下，迁移学习比现有方法具有更好的性能。

著录项

来源
《2019 Spring Simulation Conference》|2019年|1-12|共12页
会议地点 Tucson(US)
作者
Haben Yhdego; Jiang Li; Steven Morrison; Michel Audette; Christopher Paolini; Mahasweta Sarkar; Hamid Okhravi;
展开▼
作者单位

Old Dominion University, Hampton Boulevard Norfolk, VA, 5115, USA;

Old Dominion University, Hampton Boulevard Norfolk, VA, 5115, USA;

Old Dominion University, Hampton Boulevard Norfolk, VA, 5115, USA;

Old Dominion University, Hampton Boulevard Norfolk, VA, 5115, USA;

San Diego State University, Campanile Drive, San Diego, CA, 5500, USA;

San Diego State University, Campanile Drive, San Diego, CA, 5500, USA;

Eastern Virginia Medical School, Fairfax Ave, Norfolk VA, 825, USA;

展开▼
会议组织
原文格式 PDF
正文语种 eng
中图分类
关键词
Acceleration; Feature extraction; Machine learning; Continuous wavelet transforms; Accelerometers; Data models; Geriatrics;

机译：加速度;特征提取;机器学习;连续小波变换;加速度计;数据模型;老年医学;;

相似文献

外文文献
中文文献
专利

1. Deep Learning for Fall Detection: Three-Dimensional CNN Combined With LSTM on Video Kinematic Data [J] . Lu Na, Wu Yidan, Feng Li, Biomedical and Health Informatics, IEEE Journal of . 2019,第1期

机译：用于跌倒检测的深度学习：三维CNN与LSTM结合的视频运动学数据
2. A deep learning-based approach for mitigating falls from height with computer vision: Convolutional neural network [J] . Fang Weili, Zhong Botao, Zhao Neng, Advanced engineering informatics . 2019,第JANa期

机译：基于深度学习的计算机视觉缓解跌落的方法：卷积神经网络
3. Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning [J] . Hoo-Chang Shin, Holger R. Roth, Mingchen Gao, IEEE Transactions on Medical Imaging . 2016,第5期

机译：用于计算机辅助检测的深度卷积神经网络：CNN架构，数据集特征和转移学习
4. Towards Musculoskeletal Simulation-Aware Fall Injury Mitigation: Transfer Learning with Deep CNN for Fall Detection [C] . Haben Yhdego, Jiang Li, Steven Morrison, Spring Simulation Conference . 2019

机译：朝向肌肉骨骼模拟感知的秋季伤害减缓：用深CNN转移学习进行崩溃检测
5. A Deep-Learning-Based Fall-Detection System to Support Aging-in-Place. [D] . Alkittawi, Hend. 2017

机译：基于深度学习的跌落检测系统，支持就地老化。
6. Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures Dataset Characteristics and Transfer Learning [O] . Shin Hoo-Chang, Holger R. Roth, Mingchen Gao, -1

机译：用于计算机辅助检测的深度卷积神经网络：CNN架构数据集特征和转移学习
7. Deep Learning Approach Combining Lightweight CNN Architecture with Transfer Learning: An Automatic Approach for the Detection and Recognition of Bangladeshi Banknotes [O] . Ali Hasan Md. Linkon, Md. Mahir Labib, Faisal Haque Bappy, 2020

机译：轻量级CNN架构与转移学习的深度学习方法：孟加拉国钞票检测与识别的自动方法
8. Albeni Falls Wildlife Mitigation Project : 2007 Annual Report : Appendix A: Maps Showing all the Idaho Department of Fish and Game Properties under the Albeni Falls Mitigation Project. [R] . 2008

机译：阿尔贝尼瀑布野生动物减缓项目：2007年年度报告：附录a：地图显示阿尔贝尼瀑布减灾项目下的所有爱达荷鱼类和游戏物业部门。

Towards Musculoskeletal Simulation-Aware Fall Injury Mitigation: Transfer Learning with Deep CNN for Fall Detection

摘要

著录项

相似文献

相关主题

期刊订阅