Changing stroke rehab and research worldwide now.Time is Brain! trillions and trillions of neurons that DIE each day because there are NO effective hyperacute therapies besides tPA(only 12% effective). I have 523 posts on hyperacute therapy, enough for researchers to spend decades proving them out. These are my personal ideas and blog on stroke rehabilitation and stroke research. Do not attempt any of these without checking with your medical provider. Unless you join me in agitating, when you need these therapies they won't be there.

What this blog is for:

My blog is not to help survivors recover, it is to have the 10 million yearly stroke survivors light fires underneath their doctors, stroke hospitals and stroke researchers to get stroke solved. 100% recovery. The stroke medical world is completely failing at that goal, they don't even have it as a goal. Shortly after getting out of the hospital and getting NO information on the process or protocols of stroke rehabilitation and recovery I started searching on the internet and found that no other survivor received useful information. This is an attempt to cover all stroke rehabilitation information that should be readily available to survivors so they can talk with informed knowledge to their medical staff. It lays out what needs to be done to get stroke survivors closer to 100% recovery. It's quite disgusting that this information is not available from every stroke association and doctors group.

Saturday, September 18, 2021

Concurrent validity of human pose tracking in video for measuring gait parameters in older adults: a preliminary analysis with multiple trackers, viewing angles, and walking directions

 Your doctor and therapists need to use something like this to get an objective diagnosis of your gait problems. With that objective diagnosis they can then go to that database of stroke rehab protocols and select the ones that fix those disabilities.  One of my PTs said to; 'Walk this way'. Totally fucking useless no diagnosis of the problems I had and he expected me to figure out how to correct them by myself. I left him shortly after that. 

Concurrent validity of human pose tracking in video for measuring gait parameters in older adults: a preliminary analysis with multiple trackers, viewing angles, and walking directions

 

Abstract

Background

Many of the available gait monitoring technologies are expensive, require specialized expertise, are time consuming to use, and are not widely available for clinical use. The advent of video-based pose tracking provides an opportunity for inexpensive automated analysis of human walking in older adults using video cameras. However, there is a need to validate gait parameters calculated by these algorithms against gold standard methods for measuring human gait data in this population.

Methods

We compared quantitative gait variables of 11 older adults (mean age = 85.2) calculated from video recordings using three pose trackers (AlphaPose, OpenPose, Detectron) to those calculated from a 3D motion capture system. We performed comparisons for videos captured by two cameras at two different viewing angles, and viewed from the front or back. We also analyzed the data when including gait variables of individual steps of each participant or each participant’s averaged gait variables.

Results

Our findings revealed that, i) temporal (cadence and step time), but not spatial and variability gait measures (step width, estimated margin of stability, coefficient of variation of step time and width), calculated from the video pose tracking algorithms correlate significantly to that of motion capture system, and ii) there are minimal differences between the two camera heights, and walks viewed from the front or back in terms of correlation of gait variables, and iii) gait variables extracted from AlphaPose and Detectron had the highest agreement while OpenPose had the lowest agreement.

Conclusions

There are important opportunities to evaluate models capable of 3D pose estimation in video data, improve the training of pose-tracking algorithms for older adult and clinical populations, and develop video-based 3D pose trackers specifically optimized for quantitative gait measurement.

Background

Clinically established techniques for examining gait quality in older adults typically require technologies such as motion capture systems which are expensive and time consuming, require specialized expertise and staff to operate, and are not widely available for clinical use. As a result, gait monitoring practices have mainly involved cross-sectional gait assessments in laboratory settings or under experimental conditions which do not reflect the cognitive and physical demands of natural walking or usual locomotion [1].

With the advent of commercially available depth cameras, specifically the Kinect sensor (Microsoft, Redmond, WA), researchers were able to monitor natural walking of participants [2,3,4,5,6,7]. However, the Kinect camera has a limited depth of field (0.5 to 4.5 m) which can only capture few steps. This limitation, along with concerns about cost and potential hardware obsolescence (the sensor was commercially unavailable for an extended period until a newer version was released) motivate adopting other technologies for the purpose of natural gait monitoring. Although other depth sensing cameras are available, it would be ideal if technologies can make use of regular videos from cameras that are ubiquitous, such as surveillance cameras.

Advances in computer vision technology and human pose estimation in image/video data can address these limitations. A number of algorithms have been developed for human pose tracking that are capable of automated analysis of human walking using only standard RGB camera videos [8,9,10,11,12,13,14,15]. These algorithms use deep learning models that are trained on a large corpus of annotated videos, resulting in models capable of detecting body segments (head, hands, knees, feet, etc.) in new videos outside of the training dataset. These packages are freely available and can be used to process videos of human walking in any setting with minimal cost and technical expertise [15]. Gait parameters can subsequently be computed from the sequence of tracked body parts [16]. However, for use in clinical applications, there is a need to validate gait variables calculated from pose tracking data against gold standard methods for measuring human gait data, e.g., three-dimensional (3D) motion capture systems [15].

Previous studies on the validation of video pose tracking algorithms mainly used a single pose tracking algorithm, mainly OpenPose [8], in sagittal view, and healthy young adults [9, 11, 12, 15, 17]. Less is known about the performance of other publicly available pose tracking algorithms such as AlphaPose [13] or Detectron [14] particularly for pose tracking of gait in a frontal view and in older adult populations. There are several reasons that this analysis is valuable and necessary: i) comparison of different pose trackers allows researchers to choose the most appropriate one for the purpose, ii) recording walks in a frontal view allows the capture of more steps and an analysis of stability in the frontal plane, and iii) pose tracking algorithms require validation in older adults as their posture and gait are different to that in young adults and is characterized by lower speed, and greater variability [18].

The aim of this study was, therefore, to investigate the concurrent validity of spatiotemporal gait measurement in the frontal plane based on three common pose trackers (AlphaPose [13], OpenPose [8], and Detectron [14]) against a 3D motion capture system by doing a correlation analysis between the gait variables calculated from the two systems in older adults.

More at link.

No comments:

Post a Comment