Domain-Adaptive Fall Detection Using Deep Adversarial Training

Fall detection (FD) systems are important assistive technologies for healthcare that can detect emergency fall events and alert caregivers. However, it is not easy to obtain large-scale annotated fall events with various specifications of sensors or sensor positions during the implementation of accurate FD systems. Moreover, the knowledge obtained through machine learning has been restricted to tasks in the same domain. The mismatch between different domains might hinder the performance of FD systems. Cross-domain knowledge transfer is very beneficial for machine-learning based FD systems to train a reliable FD model with well-labeled data in new environments. In this study, we propose domain-adaptive fall detection (DAFD) using deep adversarial training (DAT) to tackle cross-domain problems, such as cross-position and cross-configuration. The proposed DAFD can transfer knowledge from the source domain to the target domain by minimizing the domain discrepancy to avoid mismatch problems. The experimental results show that the average F1-score improvement when using DAFD ranges from 1.5% to 7% in the cross-position scenario, and from 3.5% to 12% in the cross-configuration scenario, compared to using the conventional FD model without domain adaptation training. The results demonstrate that the proposed DAFD successfully helps to deal with cross-domain problems and to achieve better detection performance.

[1]  Klaus Hauer,et al.  The FARSEEING real-world fall repository: a large-scale collaborative database to collect and share sensor signals from real-world falls , 2016, European Review of Aging and Physical Activity.

[2]  Venkatesh Umaashankar,et al.  ActiveHARNet: Towards On-Device Deep Bayesian Active Learning for Human Activity Recognition , 2019, EMDL '19.

[3]  Reza Malekian,et al.  Fall detection monitoring systems: a comprehensive review , 2018, J. Ambient Intell. Humaniz. Comput..

[4]  Alexander D. Poularikas,et al.  The handbook of formulas and tables for signal processing , 1998 .

[5]  M. Domínguez-Morales,et al.  Wearable Fall Detector Using Recurrent Neural Networks , 2019, Sensors.

[6]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[7]  Tao Xiang,et al.  Learning to Compare: Relation Network for Few-Shot Learning , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[8]  Nigel H. Lovell,et al.  Selecting Power-Efficient Signal Features for a Low-Power Fall Detector , 2017, IEEE Transactions on Biomedical Engineering.

[9]  Timo Sztyler,et al.  Beyond position-awareness - Extending a self-adaptive fall detection system , 2019, Pervasive Mob. Comput..

[10]  Kai-Chun Liu,et al.  An Analysis of Segmentation Approaches and Window Sizes in Wearable-Based Critical Fall Detection Systems With Machine Learning Models , 2020, IEEE Sensors Journal.

[11]  Greg Mori,et al.  A comparison of accuracy of fall detection algorithms (threshold-based vs. machine learning) using waist-mounted tri-axial accelerometer signals from a comprehensive set of falls and non-fall trials , 2016, Medical & Biological Engineering & Computing.

[12]  Chao Yang,et al.  A Survey on Deep Transfer Learning , 2018, ICANN.

[13]  Roozbeh Jafari,et al.  Transferring Activity Recognition Models for New Wearable Sensors with Deep Generative Domain Adaptation , 2019, 2019 18th ACM/IEEE International Conference on Information Processing in Sensor Networks (IPSN).

[14]  A. Bourke,et al.  A threshold-based fall-detection algorithm using a bi-axial gyroscope sensor. , 2008, Medical engineering & physics.

[15]  Eduardo Casilari-Pérez,et al.  UMAFall: A Multisensor Dataset for the Research on Automatic Fall Detection , 2017, FNC/MobiSPC.

[16]  María de Lourdes Martínez-Villaseñor,et al.  UP-Fall Detection Dataset: A Multimodal Approach , 2019, Sensors.

[17]  Kai-Chun Liu,et al.  Impact of Sampling Rate on Wearable-Based Fall Detection Systems Based on Machine Learning Models , 2018, IEEE Sensors Journal.

[18]  Geoffrey E. Hinton,et al.  Visualizing Data using t-SNE , 2008 .

[19]  Eduardo Casilari-Pérez,et al.  A cross-dataset deep learning-based classifier for people fall detection and identification , 2019, Comput. Methods Programs Biomed..

[20]  Elisson Rocha,et al.  Accelerometer-Based Human Fall Detection Using Convolutional Neural Networks , 2019, Sensors.

[21]  Chin-Hui Lee,et al.  A Cross-Task Transfer Learning Approach to Adapting Deep Speech Enhancement Models to Unseen Background Noise Using Paired Senone Classifiers , 2020, ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[22]  Yu Tsao,et al.  Noise Adaptive Speech Enhancement using Domain Adversarial Training , 2018, INTERSPEECH.

[23]  Aaron Chadha,et al.  Improved Techniques for Adversarial Discriminative Domain Adaptation , 2020, IEEE Transactions on Image Processing.

[24]  Jin-Shyan Lee,et al.  Development of an Enhanced Threshold-Based Fall Detection System Using Smartphones With Built-In Accelerometers , 2019, IEEE Sensors Journal.

[25]  Regina Barzilay,et al.  Aspect-augmented Adversarial Networks for Domain Adaptation , 2017, TACL.

[26]  Sebastian Ruder,et al.  An Overview of Multi-Task Learning in Deep Neural Networks , 2017, ArXiv.

[27]  Billur Barshan,et al.  Detecting Falls with Wearable Sensors Using Machine Learning Techniques , 2014, Sensors.

[28]  Hisashi Kawai,et al.  Unsupervised Neural Adaptation Model Based on Optimal Transport for Spoken Language Identification , 2020, ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[29]  Lih-Jen Kau,et al.  A Smart Phone-Based Pocket Fall Accident Detection, Positioning, and Rescue System , 2015, IEEE Journal of Biomedical and Health Informatics.

[30]  Eduardo Casilari-Pérez,et al.  Analysis of Public Datasets for Wearable Fall Detection Systems , 2017, Sensors.

[31]  Hsinchun Chen,et al.  Hidden Markov Model-Based Fall Detection With Motion Sensor Orientation Calibration: A Case for Real-Life Home Monitoring , 2018, IEEE Journal of Biomedical and Health Informatics.

[32]  Jaime S. Cardoso,et al.  Automated Development of Custom Fall Detectors: Position, Model and Rate Impact in Performance , 2020, IEEE Sensors Journal.

[33]  Ruiyu Liang,et al.  A Deep Adaptation Network for Speech Enhancement: Combining a Relativistic Discriminator With Multi-Kernel Maximum Mean Discrepancy , 2021, IEEE/ACM Transactions on Audio, Speech, and Language Processing.

[34]  Dong Seog Han,et al.  Feature Representation and Data Augmentation for Human Activity Classification Based on Wearable IMU Sensor Data Using a Deep LSTM Neural Network , 2018, Sensors.

[35]  Haizhou Li,et al.  Unsupervised Domain Adaptation via Domain Adversarial Training for Speaker Recognition , 2018, 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[36]  Kai-Chun Liu,et al.  Novel Hierarchical Fall Detection Algorithm Using a Multiphase Fall Model , 2017, Sensors.

[37]  François Laviolette,et al.  Domain-Adversarial Training of Neural Networks , 2015, J. Mach. Learn. Res..

[38]  Maury A. Nussbaum,et al.  Preferred Placement and Usability of a Smart Textile System vs. Inertial Measurement Units for Activity Monitoring , 2018, Sensors.

[39]  Katarina Grolinger,et al.  Deep Neural Networks for Human Activity Recognition With Wearable Sensors: Leave-One-Subject-Out Cross-Validation for Model Selection , 2020, IEEE Access.

[40]  Shih-Hau Fang,et al.  Developing a mobile phone-based fall detection system on Android platform , 2012, 2012 Computing, Communications and Applications Conference.