Extraction of gait parameters from marker-free video recordings of Timed Up-and-Go tests: Validity, inter- and intra-rater reliability

Digital assessment of walking ability: Validity and reliability of the automated figure-of-eight walk test in older adults

BACKGROUND

The Figure-of-Eight Walk Test (F8WT) can assess straight- and curved-path walking ability, but the validity and reliability of automated measurement of the F8WT using digital device has not yet been studied. The aim of this study was to verify the validity (method comparison) and test-retest reliability of the automated FW8T (aFW8T) using a digital device based on image analysis by comparing the results of the aF8WT with those of the manual F8WT (mF8WT).

METHODS

Community-dwelling older adults underwent the mF8WT performed by a physiotherapist and the aF8WT using the Digital Senior Fitness Test system. To verify the test-retest reliability, the aF8WT was administered again to a randomly selected group of participants one week after the baseline test. The intraclass correlation coefficient (ICC) and Pearson's correlation analysis were used to verify the degree of agreement between the results of and correlation between the mF8WT and aF8WT, respectively. The 95% confidence interval (CI) of the limits of agreement (LoA) was obtained using Bland-Altman analysis.

RESULTS

The analysis included 83 participants (mean age 71.6 ± 4.7 years). The participants' mF8WT and aF8WT results were 29.1 ± 4.9 and 29.8 ± 4.9 seconds, respectively. Pearson's correlation analysis showed a very strong correlation between the mF8WT and aF8WT results with r = 0.91 (p < 0.001), and the ICC between the mF8WT and aF8WT results was 0.95 (0.91-0.97), showing excellent agreement. The 95% CI of the LoA was -0.7 (-4.8 to 3.3) seconds in the Bland-Altman analysis. In an analysis of the test-retest reliability of the aF8WT, participants' aF8WT results were 30.9 ± 4.7 seconds (baseline) and 29.6 ± 4.9 seconds (retest), with an ICC of 0.94 (0.81-0.98, p < 0.001), indicating excellent reliability.

CONCLUSION

Automated measurement of the F8WT using a digital device showed excellent validity and reliability. The aF8WT can be used to assess and monitor the walking ability of community-dwelling older adults.

Two-step deep-learning identification of heel keypoints from video-recorded gait

Accurate and fast extraction of step parameters from video recordings of gait allows for richer information to be obtained from clinical tests such as Timed Up and Go. Current deep-learning methods are promising, but lack in accuracy for many clinical use cases. Extracting step parameters will often depend on extracted landmarks (keypoints) on the feet. We hypothesize that such keypoints can be determined with an accuracy relevant for clinical practice from video recordings by combining an existing general-purpose pose estimation method (OpenPose) with custom convolutional neural networks (convnets) specifically trained to identify keypoints on the heel. The combined method finds keypoints on the posterior and lateral aspects of the heel of the foot in side-view and frontal-view images from which step length and step width can be determined for calibrated cameras. Six different candidate convnets were evaluated, combining three different standard architectures as networks for feature extraction (backbone), and with two different networks for predicting keypoints on the heel (head networks). Using transfer learning, the backbone networks were pre-trained on the ImageNet dataset, and the combined networks (backbone + head) were fine-tuned on data from 184 trials of older, unimpaired adults. The data was recorded at three different locations and consisted of 193 k side-view images and 110 k frontal-view images. We evaluated the six different models using the absolute distance on the floor between predicted keypoints and manually labelled keypoints. For the best-performing convnet, the median error was 0.55 cm and the 75% quartile was below 1.26 cm using data from the side-view camera. The predictions are overall accurate, but show some outliers. The results indicate potential for future clinical use by automating a key step in marker-less gait parameter extraction.

AI in evaluating ambulation of stroke patients: severity classification with video and functional ambulation category scale

BACKGROUND

The evaluation of gait function and severity classification of stroke patients are important to determine the rehabilitation goal and the level of exercise. Physicians often qualitatively evaluate patients' walking ability through visual gait analysis using naked eye, video images, or standardized assessment tools. Gait evaluation through observation relies on the doctor's empirical judgment, potentially introducing subjective opinions. Therefore, conducting research to establish a basis for more objective judgment is crucial.

OBJECTIVE

To verify a deep learning model that classifies gait image data of stroke patients according to Functional Ambulation Category (FAC) scale.

METHODS

Gait vision data from 203 stroke patients and 182 healthy individuals recruited from six medical institutions were collected to train a deep learning model for classifying gait severity in stroke patients. The recorded videos were processed using OpenPose. The dataset was randomly split into 80% for training and 20% for testing.

RESULTS

The deep learning model attained a training accuracy of 0.981 and test accuracy of 0.903. Area Under the Curve(AUC) values of 0.93, 0.95, and 0.96 for discriminating among the mild, moderate, and severe stroke groups, respectively.

CONCLUSION

This confirms the potential of utilizing human posture estimation based on vision data not only to develop gait parameter models but also to develop models to classify severity according to the FAC criteria used by physicians. To develop an AI-based severity classification model, a large amount and variety of data is necessary and data collected in non-standardized real environments, not in laboratories, can also be used meaningfully.

Effects of physical exercise interventions on balance, postural stability and general mobility in Parkinson's disease: a network meta-analysis

OBJECTIVE

To assess which type of physical exercise intervention has the most beneficial effects on balance, postural stability and general mobility in patients with Parkinson's disease. These parameters were assessed using the Activities-specific Balance Confidence (ABC) scale, Berg Balance Scale (BBS), Mini-Balance Evaluation Systems Test (MiniBESTest) and Timed Up and Go Test (TUG).

DESIGN

Network meta-analysis.

METHODS

The PubMed, Cochrane Central Register of Controlled Trials, and Web of Science databases were searched up to August 2022 to identify randomized controlled trials on the effects of physical exercise interventions on balance, postural stability, and general mobility. The network meta-analysis included pairwise and indirect comparisons of results on the ABC scale, BBS, MiniBESTest, and TUG across 8 categories of physical exercise.

RESULTS

Eighty-six studies with a total of 4,693 patients were included. For the ABC scale, the indirect comparison showed that the highest effect size was observed for balance vs sensorimotor training without including endurance interventions (0.62; 95% confidence interval (95% CI) 0.06, 1.17). The highest effect sizes for BBS were observed for alternative exercises (1.21; 95% CI 0.62, 1.81), body-weight supported (BWS) interventions (1.31; 95% CI 0.57, 2.05), dance (1.18; 95% CI 0.33, 2.03) and sensorimotor training, including endurance interventions (1.10; 95% CI 0.46, 1.75) vs control groups. Indirect comparisons showed that the highest effect size for the MiniBESTest were observed for balance (0.75; 95% CI 0.46, 1.04) and resistance (0.58; 95% CI 0.10, 1.07) vs control groups. For the TUG, comparisons showed a significant effect size for alternative exercises (-0.54; 95% CI -0.82, -0.26), balance (-0.42; 95% CI -0.75, -0.08), resistance (-0.60; 95% CI -0.89, -0.31), and sensorimotor training including endurance interventions (-0.61; 95% CI -0.95, -0.27) vs control comparisons.

CONCLUSION

Balance interventions improve balance, postural stability, and general mobility in people with Parkinson's disease. Moreover, alternative exercises, dance, BWS interventions, resistance, and sensorimotor training, including and not including endurance interventions, are also effective.

Effective evaluation of HGcnMLP method for markerless 3D pose estimation of musculoskeletal diseases patients based on smartphone monocular video

Markerless pose estimation based on computer vision provides a simpler and cheaper alternative to human motion capture, with great potential for clinical diagnosis and remote rehabilitation assessment. Currently, the markerless 3D pose estimation is mainly based on multi-view technology, while the more promising single-view technology has defects such as low accuracy and reliability, which seriously limits clinical application. This study proposes a high-resolution graph convolutional multilayer perception (HGcnMLP) human 3D pose estimation framework for smartphone monocular videos and estimates 15 healthy adults and 12 patients with musculoskeletal disorders (sarcopenia and osteoarthritis) gait spatiotemporal, knee angle, and center-of-mass (COM) velocity parameters, etc., and compared with the VICON gold standard system. The results show that most of the calculated parameters have excellent reliability (VICON, ICC (2, k): 0.853-0.982; Phone, ICC (2, k): 0.839-0.975) and validity (Pearson r: 0.808-0.978, p< 0.05). In addition, the proposed system can better evaluate human gait balance ability, and the K-means++ clustering algorithm can successfully distinguish patients into different recovery level groups. This study verifies the potential of a single smartphone video for 3D human pose estimation for rehabilitation auxiliary diagnosis and balance level recognition, and is an effective attempt at the clinical application of emerging computer vision technology. In the future, it is hoped that the corresponding smartphone program will be developed to provide a low-cost, effective, and simple new tool for remote monitoring and rehabilitation assessment of patients.

Prediction of conversion to dementia disorders based on timed up and go dual-task test verbal and motor outcomes: a five-year prospective memory-clinic-based study

BACKGROUND

While assessment tools can increase the detection of cognitive impairment, there is currently insufficient evidence regarding clinical outcomes based on screening for cognitive impairment in older adults.

METHODS

The study purpose was to investigate whether Timed Up and Go dual-task test (TUGdt) results, based on TUG combined with two different verbal tasks (name different animals, TUGdt-NA, and recite months in reverse order, TUGdt-MB), predicted dementia incidence over a period of five years among patients (N = 186, mean = 70.7 years; 45.7% female) diagnosed with Subjective Cognitive Impairment (SCI) and Mild Cognitive Impairment (MCI) following assessment at two memory clinics. Associations between TUG parameters and dementia incidence were examined in Cox regression models.

RESULTS

During follow-up time (median (range) 3.7 (0.1-6.1) years) 98 participants converted to dementia. Novel findings indicated that the TUGdt parameter words/time, after adjustment for age, gender, and education, can be used for the prediction of conversion to dementia in participants with SCI or MCI over a period of five years. Among the TUG-related parameters investigated, words/time showed the best predictive capacity, while time scores of TUG and TUGdt as well as TUGdt cost did not produce significant predictive results. Results further showed that the step parameter step length during TUGdt predicts conversion to dementia before adjustment for age, gender, and education. Optimal TUGdt cutoffs for predicting dementia at 2- and 4-year follow-up based on words/time were calculated. The sensitivity of the TUGdt cutoffs was high at 2-year follow-up: TUGdt-NA words/time, 0.79; TUGdt-MB words/time, 0.71; reducing respectively to 0.64 and 0.65 at 4-year follow-up.

CONCLUSIONS

TUGdt words/time parameters have potential as cost-efficient tools for conversion-to-dementia risk assessment, useful for research and clinical purposes. These parameters may be able to bridge the gap of insufficient evidence for such clinical outcomes.

TRIAL REGISTRATION

ClinicalTrials.gov Identifier: NCT05893524: https://www.

CLINICALTRIALS

gov/study/NCT05893524?id=NCT05893524&rank=1 .

Enabling Gait Analysis in the Telemedicine Practice through Portable and Accurate 3D Human Pose Estimation

Human pose estimation (HPE) through deep learning-based software applications is a trend topic for markerless motion analysis. Thanks to the accuracy of the state-of-the-art technology, HPE could enable gait analysis in the telemedicine practice. On the other hand, delivering such a service at a distance requires the system to satisfy multiple and different constraints like accuracy, portability, real-time, and privacy compliance at the same time. Existing solutions either guarantee accuracy and real-time (e.g., the widespread OpenPose software on well-equipped computing platforms) or portability and data privacy (e.g., light convolutional neural networks on mobile phones). We propose a portable and low-cost platform that implements real-time and accurate 3D HPE through an embedded software on a low-power off-the-shelf computing device that guarantees privacy by default and by design. We present an extended evaluation of both accuracy and performance of the proposed solution conducted with a marker-based motion capture system (i.e., Vicon) as ground truth. The results show that the platform achieves real-time performance and high-accuracy with a deviation below the error tolerance when compared to the marker-based motion capture system (e.g., less than an error of 5^∘ on the estimated knee flexion difference on the entire gait cycle and correlation 0.91<ρ<0.99). We provide a proof-of-concept study, showing that such portable technology, considering the limited discrepancies with respect to the marker-based motion capture system and its working tolerance, could be used for gait analysis at a distance without leading to different clinical interpretation.

Recognition of Freezing of Gait in Parkinson’s Disease Based on Machine Vision

Background

Freezing of gait (FOG) is a common clinical manifestation of Parkinson’s disease (PD), mostly occurring in the intermediate and advanced stages. FOG is likely to cause patients to fall, resulting in fractures, disabilities and even death. Currently, the pathogenesis of FOG is unclear, and FOG detection and screening methods have various defects, including subjectivity, inconvenience, and high cost. Due to limited public healthcare and transportation resources during the COVID-19 pandemic, there are greater inconveniences for PD patients who need diagnosis and treatment.

Objective

A method was established to automatically recognize FOG in PD patients through videos taken by mobile phone, which is time-saving, labor-saving, and low-cost for daily use, which may overcome the above defects. In the future, PD patients can undergo FOG assessment at any time in the home rather than in the hospital.

Methods

In this study, motion features were extracted from timed up and go (TUG) test and the narrow TUG (Narrow) test videos of 50 FOG-PD subjects through a machine learning method; then a motion recognition model to distinguish between walking and turning stages and a model to recognize FOG in these stages were constructed using the XGBoost algorithm. Finally, we combined these three models to form a multi-stage FOG recognition model.

Results

We adopted the leave-one-subject-out (LOSO) method to evaluate model performance, and the multi-stage FOG recognition model achieved a sensitivity of 87.5% sensitivity and a specificity of 79.82%.

Conclusion

A method to realize remote PD patient FOG recognition based on mobile phone video is presented in this paper. This method is convenient with high recognition accuracy and can be used to rapidly evaluate FOG in the home environment and remotely manage FOG-PD, or screen patients in large-scale communities.