Multi-Level Adversarial Spatio-Temporal Learning for Footstep Pressure Based FoG Detection
文献类型:期刊论文
作者 | Hu, Kun4; Mei, Shaohui5; Wang, Wei1,6; Martens, Kaylena A. Ehgoetz2; Wang, Liang1,6,7; Lewis, Simon J. G.3; Feng, David D.4; Wang, Zhiyong4 |
刊名 | IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS |
出版日期 | 2023-08-01 |
卷号 | 27期号:8页码:4166-4177 |
ISSN号 | 2168-2194 |
关键词 | Adversarial learning deep learning freezing of gait detection footstep pressure parkinson's disease |
DOI | 10.1109/JBHI.2023.3272902 |
通讯作者 | Hu, Kun(kun.hu@sydney.edu.au) |
英文摘要 | Freezing of gait (FoG) is one of the most common symptoms of Parkinson's disease, which is a neurodegenerative disorder of the central nervous system impacting millions of people around the world. To address the pressing need to improve the quality of treatment for FoG, devising a computer-aided detection and quantification tool for FoG has been increasingly important. As a non-invasive technique for collecting motion patterns, the footstep pressure sequences obtained from pressure sensitive gait mats provide a great opportunity for evaluating FoG in the clinic and potentially in the home environment. In this study, FoG detection is formulated as a sequential modelling task and a novel deep learning architecture, namely Adversarial Spatio-temporal Network (ASTN), is proposed to learn FoG patterns across multiple levels. ASTN introduces a novel adversarial training scheme with a multi-level subject discriminator to obtain subject-independent FoG representations, which helps to reduce the over-fitting risk due to the high inter-subject variance. As a result, robust FoG detection can be achieved for unseen subjects. The proposed scheme also sheds light on improving subject-level clinical studies from other scenarios as it can be integrated with many existing deep architectures. To the best of our knowledge, this is one of the first studies of footstep pressure-based FoG detection and the approach of utilizing ASTN is the first deep neural network architecture in pursuit of subject-independent representations. In our experiments on 393 trials collected from 21 subjects, the proposed ASTN achieved an AUC 0.85, clearly outperforming conventional learning methods. |
WOS关键词 | PARKINSONS-DISEASE ; GAIT ; MOBILE ; FALLS |
资助项目 | Australian Research Council (ARC)[DP210102674] ; Australian Research Council (ARC)[DP160103675] ; NHMRC-ARC Dementia Fellowship[1110414] ; National Health and Medical Research Council (NHMRC) of Australia Program[1037746] ; Dementia Research Team[1095127] ; Neuro Sleep Centre of Research Excellence[1060992] ; ARC Centre of Excellence in Cognition and Its Disorders Memory Program[CE110001021] ; Sydney Research Excellence Initiative (SREI) 2020 of the University of Sydney ; Natural Science Foundation of China[61420106015] ; Parkinson Canada ; Brain and Mind Centre Early Career Research Development Grant |
WOS研究方向 | Computer Science ; Mathematical & Computational Biology ; Medical Informatics |
语种 | 英语 |
出版者 | IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC |
WOS记录号 | WOS:001045824200042 |
资助机构 | Australian Research Council (ARC) ; NHMRC-ARC Dementia Fellowship ; National Health and Medical Research Council (NHMRC) of Australia Program ; Dementia Research Team ; Neuro Sleep Centre of Research Excellence ; ARC Centre of Excellence in Cognition and Its Disorders Memory Program ; Sydney Research Excellence Initiative (SREI) 2020 of the University of Sydney ; Natural Science Foundation of China ; Parkinson Canada ; Brain and Mind Centre Early Career Research Development Grant |
源URL | [http://ir.ia.ac.cn/handle/173211/53996] |
专题 | 多模态人工智能系统全国重点实验室 |
通讯作者 | Hu, Kun |
作者单位 | 1.Univ Chinese Acad Sci UCAS, Beijing 100040, Peoples R China 2.Univ Waterloo, Waterloo, ON N2L 3G1, Canada 3.Univ Sydney, Brain & Mind Ctr, ForeFront Parkinsons Dis Res Clin, Sydney, NSW 2050, Australia 4.Univ Sydney, Sch Comp Sci, Sydney, NSW 2006, Australia 5.Northwestern Polytech Univ, Sch Elect & Informat, Xian 710072, Peoples R China 6.Chinese Acad Sci CASIA, Inst Automat, Ctr Res Intelligent Percept & Comp CRIPAC, Natl Lab Pattern Recognit NLPR, Beijing 100190, Peoples R China 7.Chinese Acad Sci CASIA, Inst Automat, Ctr Excellence Brain Sci & Intelligence Technol CE, Beijing 100190, Peoples R China |
推荐引用方式 GB/T 7714 | Hu, Kun,Mei, Shaohui,Wang, Wei,et al. Multi-Level Adversarial Spatio-Temporal Learning for Footstep Pressure Based FoG Detection[J]. IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS,2023,27(8):4166-4177. |
APA | Hu, Kun.,Mei, Shaohui.,Wang, Wei.,Martens, Kaylena A. Ehgoetz.,Wang, Liang.,...&Wang, Zhiyong.(2023).Multi-Level Adversarial Spatio-Temporal Learning for Footstep Pressure Based FoG Detection.IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS,27(8),4166-4177. |
MLA | Hu, Kun,et al."Multi-Level Adversarial Spatio-Temporal Learning for Footstep Pressure Based FoG Detection".IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS 27.8(2023):4166-4177. |
入库方式: OAI收割
来源:自动化研究所
浏览0
下载0
收藏0
其他版本
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。