Measuring Gait Variables Using Computer Vision to Assess Mobility and Fall Risk in Older Adults With Dementia

Fall risk is high for older adults with dementia. Gait impairment contributes to increased fall risk, and gait changes are common in people with dementia, although the reliable assessment of gait is challenging in this population. This study aimed to develop an automated approach to performing gait assessments based on gait data that is collected frequently and unobtrusively, and analysed using computer vision methods. Recent developments in computer vision have led to the availability of open source human pose estimation algorithms, which automatically estimate the joint locations of a person in an image. In this study, a pre-existing pose estimation model was applied to 1066 walking videos collected of 31 older adults with dementia as they walked naturally in a corridor on a specialized dementia unit over a two week period. Using the tracked pose information, gait features were extracted from video recordings of gait bouts and their association with clinical mobility assessment scores and future falls data was examined. A significant association was found between extracted gait features and a clinical mobility assessment and the number of future falls, providing concurrent and predictive validation of this approach.

[1]  Lourdes Agapito,et al.  Lifting from the Deep: Convolutional 3D Pose Estimation from a Single Image , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[2]  Chang Oh Kim,et al.  Quantitative analysis of the bilateral coordination and gait asymmetry using inertial measurement unit-based gait analysis , 2019, PloS one.

[3]  Avril Mansfield,et al.  Does the margin of stability measure predict medio-lateral stability of gait with a constrained-width base of support? , 2019, Journal of biomechanics.

[4]  Chen Gong,et al.  Automatic Timed Up-and-Go Sub-Task Segmentation for Parkinson’s Disease Patients Using Video-Based Activity Classification , 2018, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[5]  K. Musselman,et al.  Evaluating Mobility in Advanced Dementia: A Scoping Review and Feasibility Analysis. , 2018, The Gerontologist.

[6]  J. Morley,et al.  Gait, Falls, and Dementia. , 2016, Journal of the American Medical Directors Association.

[7]  Guang-Zhong Yang,et al.  3-D Canonical Pose Estimation and Abnormal Gait Recognition With a Single RGB-D Camera , 2019, IEEE Robotics and Automation Letters.

[8]  A. Hof.  The 'extrapolated center of mass' concept suggests a simple control of balance in walking. , 2008, Human movement science.

[9]  Babak Taati,et al.  Quantitative Mobility Assessment for Fall Risk Prediction in Dementia: A Systematic Review , 2018, Dementia and Geriatric Cognitive Disorders.

[10]  Yichen Wei,et al.  Simple Baselines for Human Pose Estimation and Tracking , 2018, ECCV.

[11]  Andrew W. Fitzgibbon,et al.  Real-time human pose recognition in parts from single depth images , 2011, CVPR 2011.

[12]  Yaser Sheikh,et al.  OpenPose: Realtime Multi-Person 2D Pose Estimation Using Part Affinity Fields , 2018, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[13]  Xiaowei Zhou,et al.  MonoCap: Monocular Human Motion Capture using a CNN Coupled with a Geometric Prior , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[14]  Varun Ramakrishna,et al.  Convolutional Pose Machines , 2016, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[15]  M. Hannan,et al.  EVALUATING GAIT SYMMETRY AND LEG DOMINANCE DURING WALKING IN HEALTHY OLDER ADULTS , 2011 .

[16]  Yinghao Huang,et al.  Towards Accurate Marker-Less Human Shape and Pose Estimation over Time , 2017, 2017 International Conference on 3D Vision (3DV).

[17]  Li Fei-Fei,et al.  Vision-Based Gait Analysis for Senior Care , 2018, ArXiv.

[18]  P. Slattum,et al.  The Cost of Falls Among the Community-Dwelling Elderly , 2005, Journal of managed care pharmacy : JMCP.

[19]  Babak Taati,et al.  The feasibility of a vision-based sensor for longitudinal monitoring of mobility in older adults with dementia. , 2019, Archives of gerontology and geriatrics.

[20]  E. Camargos,et al.  Wrist actigraphy for measuring sleep in intervention studies with Alzheimer's disease patients: application, usefulness, and challenges. , 2013, Sleep medicine reviews.

[21]  Julia Richter,et al.  Pose Estimation and Movement Detection for Mobility Assessment of Elderly People in an Ambient Assisted Living Application , 2015, ICPRAM.

[22]  Tao Liu,et al.  Wearable Sensor System for Detecting Gait Parameters of Abnormal Gaits: A Feasibility Study , 2018, IEEE Sensors Journal.

[23]  Torsten Bumgarner,et al.  Biomechanics and Motor Control of Human Movement , 2013 .

[24]  Jia Deng,et al.  Stacked Hourglass Networks for Human Pose Estimation , 2016, ECCV.

[25]  Yaser Sheikh,et al.  OpenPose: Realtime Multi-Person 2D Pose Estimation Using Part Affinity Fields , 2018, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[26]  Alex Mihailidis,et al.  Vision-based approach for long-term mobility monitoring: Single case study following total hip replacement. , 2014, Journal of rehabilitation research and development.

[27]  Babak Taati,et al.  Vision-based assessment of gait features associated with falls in people with dementia. , 2020, The journals of gerontology. Series A, Biological sciences and medical sciences.

[28]  Ahmad Lotfi,et al.  Video Based Fall Detection using Features of Motion, Shape and Histogram , 2018, PETRA.

[29]  Yang Liu,et al.  Video-based Fall Detection for Seniors with Human Pose Estimation , 2018, 2018 4th International Conference on Universal Village (UV).

[30]  Marjan J Faber,et al.  Clinimetric properties of the performance-oriented mobility assessment. , 2006, Physical therapy.

[31]  Gangolf Hirtz,et al.  Estimation of Gait Parameters from 3D Pose for Elderly Care , 2019, ICBBE.

[32]  Wiebren Zijlstra,et al.  Sensor-based assessment of mobility-related behavior in dementia: feasibility and relevance in a hospital context , 2016, International Psychogeriatrics.

[33]  Alex Mihailidis,et al.  Concurrent validity of the Microsoft Kinect for Windows v2 for measuring spatiotemporal gait parameters. , 2016, Medical engineering & physics.

[34]  Moataz Eltoukhy,et al.  Improved kinect-based spatiotemporal and kinematic treadmill gait assessment. , 2017, Gait & posture.

[35]  S L Luther,et al.  Fall risk assessment measures: an analytic review. , 2001, The journals of gerontology. Series A, Biological sciences and medical sciences.

[36]  M. Tinetti.  Performance‐Oriented Assessment of Mobility Problems in Elderly Patients , 1986, Journal of the American Geriatrics Society.

[37]  George Fulk,et al.  Test-Retest Reliability and Construct Validity of the Tinetti Performance-Oriented Mobility Assessment in People With Stroke , 2013, Journal of neurologic physical therapy : JNPT.

[38]  Kei Masani,et al.  Relationship between margin of stability and deviations in spatiotemporal gait features in healthy young adults. , 2017, Human movement science.

[39]  A L Hof,et al.  Ultrasonic motion analysis system--measurement of temporal and spatial gait parameters. , 2002, Journal of biomechanics.

[40]  Bernt Schiele,et al.  2D Human Pose Estimation: New Benchmark and State of the Art Analysis , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[41]  Hongnian Yu,et al.  An Automatic Gait Feature Extraction Method for Identifying Gait Asymmetry Using Wearable Sensors , 2018, Sensors.

[42]  Ross B. Girshick,et al.  Mask R-CNN , 2017, 1703.06870.

[43]  John K. Tsotsos,et al.  Vision-Based Fallen Person Detection for the Elderly , 2017, 2017 IEEE International Conference on Computer Vision Workshops (ICCVW).