A pilot observational study of gait changes over time before and after an unplanned hospital visit in long-term care residents with dementia.

BACKGROUND: Older adults with dementia living in long-term care (LTC) have high rates of hospitalization. Two common causes of unplanned hospital visits for LTC residents are deterioration in health status and falls. Early detection of health deterioration or increasing falls risk may present an opportunity to intervene and prevent hospitalization. There is some evidence that impairments in older adults' gait, such as reduced gait speed, increased variability, and poor balance may be associated with hospitalization. However, it is not clear whether changes in gait are observable and measurable before an unplanned hospital visit and whether these changes persist after the acute medical issue has been resolved. The objective of this study was to examine gait changes before and after an unplanned acute care hospital visit in people with dementia. METHODS: We performed a secondary analysis of quantitative gait measures extracted from videos of natural gait captured over time on a dementia care unit and collected information about unplanned hospitalization from health records. RESULTS: Gait changes in study participants before hospital visits were characterized by decreasing stability and step length, and increasing step variability, although these changes were also observed in participants without hospital visits. In an age and sex-adjusted mixed effects model, gait speed and step length declined more quickly in those with a hospital visit compared to those without. CONCLUSIONS: These results provide preliminary evidence that clinically meaningful longitudinal gait changes may be captured by repeated non-invasive gait monitoring, although a larger study is needed to identify changes specific to future medical events.

Ambient Monitoring of Gait and Machine Learning Models for Dynamic and Short-Term Falls Risk Assessment in People With Dementia.

Falls are a leading cause of morbidity and mortality in older adults with dementia residing in long-term care. Having access to a frequently updated and accurate estimate of the likelihood of a fall over a short time frame for each resident will enable care staff to provide targeted interventions to prevent falls and resulting injuries. To this end, machine learning models to estimate and frequently update the risk of a fall within the next 4 weeks were trained on longitudinal data from 54 older adult participants with dementia. Data from each participant included baseline clinical assessments of gait, mobility, and fall risk at the time of admission, daily medication intake in three medication categories, and frequent assessments of gait performed via a computer vision-based ambient monitoring system. Systematic ablations investigated the effects of various hyperparameters and feature sets and experimentally identified differential contributions from baseline clinical assessments, ambient gait analysis, and daily medication intake. In leave-one-subject-out cross-validation, the best performing model predicts the likelihood of a fall over the next 4 weeks with a sensitivity and specificity of 72.8 and 73.2, respectively, and achieved an area under the receiver operating characteristic curve (AUROC) of 76.2. By contrast, the best model excluding ambient gait features achieved an AUROC of 56.2 with a sensitivity and specificity of 51.9 and 54.0, respectively. Future research will focus on externally validating these findings to prepare for the implementation of this technology to reduce fall and fall-related injuries in long-term care.

Developing a Clinical Decision Support System for Prediction Postoperative Coronary Artery Bypass Grafting Infection in Diabetic Patients.

Background: Postoperative infection in Coronary Artery Bypass Graft (CABG) is one of the most common complications for diabetic patients, due to an increase in the hospitalization and cost. To address these issues, it is necessary to apply some solutions. Objective: The study aimed to the development of a Clinical Decision Support System (CDSS) for predicting the CABG postoperative infection in diabetic patients. Material and Methods: This developmental study is conducted on a private hospital in Tehran in 2016. From 1061 CABG surgery medical records, we selected 210 cases randomly. After data gathering, we used statistical tests for selecting related features. Then an Artificial Neural Network (ANN), which was a one-layer perceptron network model and a supervised training algorithm with gradient descent, was constructed using MATLAB software. The software was then developed and tested using the receiver operating characteristic (ROC) diagram and the confusion matrix. Results: Based on the correlation analysis, from 28 variables in the data, 20 variables had a significant relationship with infection after CABG (P<0.05). The results of the confusion matrix showed that the sensitivity of the system was 69%, and the specificity and the accuracy were 97% and 84%, respectively. The Receiver Operating Characteristic (ROC) diagram shows the appropriate performance of the CDSS. Conclusion: The use of CDSS can play an important role in predicting infection after CABG in patients with diabetes. The designed software can be used as a supporting tool for physicians to predict infections caused by CABG in diabetic patients as a susceptible group. However, other factors affecting infection must also be considered for accurate prediction.

The Toronto older adults gait archive: video and 3D inertial motion capture data of older adults' walking.

We introduce the Toronto Older Adults Gait Archive, a gait dataset of 14 older adults containing 2D video recordings, and 2D (video pose tracking algorithms) and 3D (inertial motion capture) joint locations of the lower body. Participants walked for 60 seconds. We also collected participants' scores on four clinical assessments of gait and balance, namely the Tinneti performance-oriented mobility assessment (POMA-gait and -balance), the Berg balance scale (BBS), and the timed-up-and-go (TUG). Three human pose tracking models (Alphapose, OpenPose, and Detectron) were used to detect body joint positions in 2D video frames and a number of gait parameters were computed using 2D video-based and 3D motion capture data. To show an example usage of our datasets, we performed a correlation analysis between the gait variables and the clinical scores. Our findings revealed that the temporal but not the spatial or variability gait variables from both systems had high correlations to clinical scores. This dataset can be used to evaluate, or to enhance vision-based pose-tracking models to the specifics of older adults' walking.

Estimating Parkinsonism Severity in Natural Gait Videos of Older Adults With Dementia.

Drug-induced parkinsonism affects many older adults with dementia, often causing gait disturbances. New advances in vision-based human pose- estimation have opened possibilities for frequent and unobtrusive analysis of gait in long-term care settings. This work leverages spatial-temporal graph convolutional network (ST-GCN) architectures and training procedures to predict clinical scores of parkinsonism in gait from video of individuals with dementia. We propose a two-stage training approach consisting of a self-supervised pretraining stage that encourages the ST-GCN model to learn about gait patterns before predicting clinical scores in the finetuning stage. The proposed ST-GCN models are evaluated on joint trajectories extracted from video and are compared against traditional (ordinal, linear, random forest) regression models and temporal convolutional network baselines. Three 2D human pose-estimation libraries (OpenPose, Detectron, AlphaPose) and the Microsoft Kinect (2D and 3D) are used to extract joint trajectories of 4787 natural walking bouts from 53 older adults with dementia. A subset of 399 walks from 14 participants is annotated with scores of parkinsonism severity on the gait criteria of the Unified Parkinson's Disease Rating Scale (UPDRS) and the Simpson-Angus Scale (SAS). Our results demonstrate that ST-GCN models operating on 3D joint trajectories extracted from the Kinect consistently outperform all other models and feature sets. Prediction of parkinsonism scores in natural walking bouts of unseen participants remains a challenging task, with the best models achieving macro-averaged F1-scores of 0.53 ± 0.03 and 0.40 ± 0.02 for UPDRS-gait and SAS-gait, respectively. Pre-trained model and demo code for this work is available.1.

Prediction of Parkinsonian Gait in Older Adults with Dementia using Joint Trajectories and Gait Features from 2D Video.

Older adults with dementia have a high risk of developing drug-induced parkinsonism; however, formal clinical gait assessments are too infrequent to capture fluctuations in their gait. Camera-based human pose estimation and tracking provides a means to frequently monitor gait in nonclinical settings. In this study, 2160 walking bouts from 49 participants were recorded using a ceiling-mounted camera. Recorded color videos were processed using AlphaPose to obtain 2D joint trajectories of the participant as they were walking down a hallway of the unit. A subset of 324 walking bouts from 14 participants were annotated with clinical scores of parkinsonism on the Unified Parkinson's Disease Rating Scale (UPDRS)-gait scale. Linear, random forest, and ordinal logistic regression models were evaluated for regression to UPDRS-gait scores using engineered 2D gait features calculated from the AlphaPose joint trajectories. Additionally, spatial temporal graph convolutional networks (ST-GCNs) were trained to predict UPDRS-gait scores from joint trajectories and gait features using a two-stage training scheme (self-supervised pretraining stage on all walks followed by a finetuning stage on labelled walks). All models were trained using leave-one-subject-out cross-validation to simulate testing on previously unseen participants. The macro-averaged F1-score was 0.333 for the best model operating on only gait features and 0.372 for the top ST-GCN model that used both joint trajectories and gait features as input. When accepting predicted scores that were only off by at most 1 point on the UPDRS-gait scale, the accuracy of the model that only used gait features was 82.8%, while the model that also used joint trajectories had an accuracy of 94.2%.Clinical Relevance- The combination of gait features and joint trajectories capture parkinsonian qualities in gait better than either group of data individually.

Gait changes over time in hospitalized older adults with advanced dementia: Predictors of mobility change.

People with dementia are at risk of mobility decline. In this study, we measured changes in quantitative gait measures over a maximum 10-week period during the course of a psychogeriatric admission in older adults with dementia, with the aims to describe mobility changes over the duration of the admission, and to determine which factors were associated with this change. Fifty-four individuals admitted to a specialized dementia inpatient unit participated in this study. A vision-based markerless motion capture system was used to record participants' natural gait. Mixed effect models were developed with gait measures as the dependent variables and clinical and demographic variables as predictors. We found that gait stability, step time, and step length decreased, and step time variability and step length variability increased over 10 weeks. Gait stability of men decreased more than that of women, associated with an increased sacrum mediolateral range of motion over time. In addition, the sacrum mediolateral range of motion decreased in those with mild neuropsychiatric symptoms over 10 weeks, but increased in those with more severe neuropsychiatric symptoms. Our study provides evidence of worsening of gait mechanics and control over the course of a hospitalization in older adults with dementia. Quantitative gait monitoring in hospital environments may provide opportunities to intervene to prevent adverse events, decelerate mobility decline, and monitor rehabilitation outcomes.

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.

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.

Effect of simulated sensorimotor noise on kinematic variability and stability of a biped walking model.

Whether higher variability in older adults' walking is an indication of increased instability has been challenged recently. We performed a computer simulation to investigate the effect of sensorimotor noise on the kinematic variability and stability in a biped walking model. Stochastic differential equations of the system with additive Gaussian white noise was constructed and solved. Sensorimotor noise mainly resulted in higher kinematic variability but its influence on gait stability is minimal. This implies that kinematic variability evident in walking gaits of older adults could be the result of internal sensorimotor noise and not an indication of instability.

Predicting Short-Term Risk of Falls in a High-Risk Group With Dementia.

OBJECTIVES: To develop a prognostic model to predict the probability of a short-term fall (within the next 7 to 30 days) in older adults with dementia. DESIGN: Prospective observational study. SETTING AND PARTICIPANTS: Fifty-one individuals with dementia at high risk of falls from a specialized dementia inpatient unit. METHODS: Clinical and demographic measures were collected and a vision-based markerless motion capture was used to record the natural gait of participants over a 2-week baseline. Falls were tracked throughout the length of stay. Cox proportional hazard regression analysis was used to build a prognostic model to determine fall-free survival probabilities at 7 days and at 30 days. The model's discriminative ability was also internally validated. RESULTS: Fall history and gait stability (estimated margin of stability) were statistically significant predictors of time to fall and included in the final prognostic model. The model's predicted survival probabilities were close to observed values at both 7 and 30 days. The area under the receiver operating curve was 0.80 at 7 days, and 0.67 at 30 days and the model had a discrimination performance (the Harrel concordance index) of 0.71. CONCLUSIONS AND IMPLICATIONS: Our short-term falls risk model had fair to good predictive and discrimination ability. Gait stability and recent fall history predicted an imminent fall in our population. This provides some preliminary evidence that the degree of gait instability may be measureable in natural everyday gait to allow dynamic falls risk monitoring. External validation of the model using a separate data set is needed to evaluate model's predictive performance.

A systematic review of center of pressure measures to quantify gait changes in older adults.

Measures of gait center of pressure (COP) can be recorded using simple available technologies in clinical settings and thus can be used to characterize gait quality in older adults and its relationship to falls. The aim of this systematic review was to investigate the association between measures of gait COP and aging and falls. A comprehensive search of electronic databases including MEDLINE, Embase, Cochrane Central Register of Controlled Trials, CINAHL (EBSCO), Ageline (EBSCO) and Scopus was performed. The initial search yielded 2809 papers. After removing duplicates and applying study inclusion/exclusion criteria, 34 papers were included in the review. Gait COP has been examined during three tasks: normal walking, gait initiation, and obstacle negotiation. The majority of studies examined mean COP position and velocity as outcome measures. Overall, gait in older adults was characterized by more medial COP trajectory in normal walking and lower average anterior-posterior and medio-lateral COP displacements and velocity in both gait initiation and obstacle crossing. Moreover, findings suggest that Tai chi training can enhance older adults' balance control during gait initiation as demonstrated by greater COP backward, medial and forward shift in all three phases of gait initiation. These findings should be interpreted cautiously due to inadequacy of evidence as well as methodological limitations of the studies such as small sample size, limited numbers of 'fallers', lack of a control group, and lack of interpretation of COP outcomes with respect to fall risk. COP measures can be adopted to assess fall-related gait changes in older adults but more complex measures of COP that reveal the dynamic nature of COP behavior in step-to-step variations are needed to adequately characterize gait changes in older adults.

Assessment of Parkinsonian gait in older adults with dementia via human pose tracking in video data.

BACKGROUND: Parkinsonism is common in people with dementia, and is associated with neurodegenerative and vascular changes in the brain, or with exposure to antipsychotic or other dopamine antagonist medications. The detection of parkinsonian changes to gait may provide an opportunity to intervene and address reversible causes. In this study, we investigate the use of a vision-based system as an unobtrusive means to assess severity of parkinsonism in gait. METHODS: Videos of walking bouts of natural gait were collected in a specialized dementia unit using a Microsoft Kinect sensor and onboard color camera, and were processed to extract sixteen 3D and eight 2D gait features. Univariate regression to gait quality, as rated on the Unified Parkinson's Disease Rating Scale (UPDRS) and Simpson-Angus Scale (SAS), was used to identify gait features significantly correlated to these clinical scores for inclusion in multivariate models. Multivariate ordinal logistic regression was subsequently performed and the relative contribution of each gait feature for regression to UPDRS-gait and SAS-gait scores was assessed. RESULTS: Four hundred one walking bouts from 14 older adults with dementia were included in the analysis. Multivariate ordinal logistic regression models incorporating selected 2D or 3D gait features attained similar accuracies: the UPDRS-gait regression models achieved accuracies of 61.4 and 62.1% for 2D and 3D features, respectively. Similarly, the SAS-gait models achieved accuracies of 47.4 and 48.5% with 2D or 3D gait features, respectively. CONCLUSIONS: Gait features extracted from both 2D and 3D videos are correlated to UPDRS-gait and SAS-gait scores of parkinsonism severity in gait. Vision-based systems have the potential to be used as tools for longitudinal monitoring of parkinsonism in residential settings.

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.

Vision-Based Assessment of Gait Features Associated With Falls in People With Dementia.

BACKGROUND: Gait impairments contribute to falls in people with dementia. In this study, we used a vision-based system to record episodes of walking over a 2-week period as participants moved naturally around their environment, and from these calculated spatiotemporal, stability, symmetry, and acceleration gait features. The aim of this study was to determine whether features of gait extracted from a vision-based system are associated with falls, and which of these features are most strongly associated with falling. METHODS: Fifty-two people with dementia admitted to a specialized dementia unit participated in this study. Thirty different features describing baseline gait were extracted from Kinect recordings of natural gait over a 2-week period. Baseline clinical and demographic measures were collected, and falls were tracked throughout the participants' admission. RESULTS: A total of 1,744 gait episodes were recorded (mean 33.5 ± 23.0 per participant) over a 2-week baseline period. There were a total of 78 falls during the study period (range 0-10). In single variable analyses, the estimated lateral margin of stability, step width, and step time variability were significantly associated with the number of falls during admission. In a multivariate model controlling for clinical and demographic variables, the estimated lateral margin of stability (p = .01) was remained associated with number of falls. CONCLUSIONS: Information about gait can be extracted from vision-based recordings of natural walking. In particular, the lateral margin of stability, a measure of lateral gait stability, is an important marker of short-term falls risk.

Lower extremity kinematic analysis in male athletes with unilateral anterior cruciate reconstruction in a jump-landing task and its association with return to sport criteria.

BACKGROUND: Return to sport (RTS) criteria are widely being used to identify anterior cruciate ligament reconstructed (ACLR) athletes ready to return to sportive activity and reduce risk of ACL re-injury. However, studies show a high rate of ACL re-injury in athletes who passed RTS criteria. This indicates that the current RTS criteria might not be sufficient to determine return to sport time in ACLR athletes. Previous studies have reported a close association between altered lower limb kinematics and ACL re-injury. However, it is not clear how lower extremity kinematics differs between ACLR athletes who passed the RTS-criteria and who failed. This study compared lower extremity kinematics in a jump-landing task between ACLR athletes who passed the RTS criteria (Limb symmetry in hop tests, quadriceps strength and questionnaires) to those who failed and to the healthy individuals. METHODS: Participants were 27 male football players with unilateral ACLR including 14 who passed -RTS criteria and 13 failed, and 15 healthy football players. A 3D motion capture system recorded participants' lower extremity motion while performing 10 trials of a bilateral jump-landing task. Hip, knee and ankle angular motion were examined at initial contact. Two-way mixed analysis of variances (2 limbs × 3 groups) and Bonferroni post-hoc tests were performed to compare the joint angles between the limbs and groups. RESULTS: lower hip abduction angle was found in the failed (involved limb 4.1 ° ± 4.2) and passed RTS (involved limb 6.8° ± 3.3) groups compared to the healthy group (non-dominant limb 10.7° ± 3.7). Ankle inversion in the failed RTS (0.4° ± 4.9) group was significantly lower than both passed RTS (4.8° ± 4.8, p = 0.05) and healthy (8.2° ± 8.1, p < 0.001) groups. There were no significant differences between the groups in knee kinematics. CONCLUSIONS: Our findings indicate reduced hip abduction during initial contact phase of landing in athletes returned to sport. Reduced hip abduction during the complex multiplanar movement of jump-landing is a risk factor for ACL re-injury. Current RTS criteria may not be sufficient to identify ACLR athletes at high risk of re-injury. The kinematic analysis in conjunction with current RTS criteria can provide additional insight into the return to sport decision making.

Gait stability of diabetic patients is altered with the rigid rocker shoes.

BACKGROUND: Rigid-rocker shoes may induce gait instability in diabetics, however, this is not clearly investigated. The present study investigates if rigid-rocker shoes influence diabetic gait stability. METHODS: Fourteen non-neuropathic and nine neuropathic diabetics, plus eleven healthy young-adults were recruited. Full-body kinematic data was captured during walking. Experimental conditions included barefoot and three rocker-shoe designs according to the rocker angle, apex angle and apex position (R10: 10°, 80°, 60%; R15: 15°, 95°, 52%; R20: 20°, 95°, 60%). Sagittal and frontal stability margin, plus fear of fall were main outcome measures. FINDINGS: Sagittal stability margin was not affected by health, however, was increased with R10 and R15 in non-neuropathic diabetics and healthy individuals (R2 = 0.16). Variability of sagittal stability margin was not altered in neuropathic diabetics, but was increased with R15 and R20 in healthy participants, with R15 in non-neuropathic diabetics (R2 = 0.12). Frontal stability margin (R2 = 0.46) and its variability (R2 = 0.39) were significantly increased in neuropathic and non-neuropathic diabetics compared to healthy individuals. Frontal stability margin was significantly higher with R15 in neuropathic diabetics, and with R20 in both non-neuropathic and healthy participants. Sagittal and frontal stability margin were strongly correlated with fear of fall in neuropathic diabetics. INTERPRETATIONS: R15 and R20 might challenge gait stability of diabetics cause them restrict centre of mass motion thereby imposing a tighter control over walking. However, neuropathic diabetics generally walk very cautious due to neuropathy and increased fear of fall. Frontal stability margin, highly affected by health and experimental condition, is a more sensitive indicator of gait stability.

In search of sports biomechanics' holy grail: Can athlete-specific optimum sports techniques be identified?

The development of methods that can identify athlete-specific optimum sports techniques-arguably the holy grail of sports biomechanics-is one of the greatest challenges for researchers in the field. This 'perspectives article' critically examines, from a dynamical systems theoretical standpoint, the claim that athlete-specific optimum sports techniques can be identified through biomechanical optimisation modelling. To identify athlete-specific optimum sports techniques, dynamical systems theory suggests that a representative set of organismic constraints, along with their non-linear characteristics, needs to be identified and incorporated into the mathematical model of the athlete. However, whether the athlete will be able to adopt, and reliably reproduce, his/her predicted optimum technique will largely be dependent on his/her intrinsic dynamics. If the attractor valley corresponding to the existing technique is deep, or if the attractor valleys corresponding to the existing technique and the predicted optimum technique are in different topographical regions of the dynamic landscape, technical modifications may be challenging or impossible to reliably implement even after extended practice. The attractor layout defining the intrinsic dynamics of the athlete, therefore, needs to be determined to establish the likelihood of the predicted optimum technique being reliably attainable by the athlete. Given the limited set of organismic constraints typically used in mathematical models of athletes, combined with the methodological challenges associated with mapping the attractor layout of an athlete, it seems unlikely that athlete-specific optimum sports techniques will be identifiable through biomechanical optimisation modelling for the majority of sports skills in the near future.

A robust method to estimate the largest Lyapunov exponent of noisy signals: A revision to the Rosenstein's algorithm.

This study proposed a revision to the Rosenstein's method of numerical calculation of the largest Lyapunov exponent (LyE) to make it more robust to noise. To this aim, the effect of increasing number of initial neighboring points on the LyE value was investigated and compared to values obtained by filtering the time series. Both simulated (Lorenz and passive dynamic walker) and experimental (human walking) time series were used to calculate the LyE. The number of initial neighbors used to calculate LyE for all time series was 1 (the original Rosenstein's method), 2, 3, 4, 5, 10, 15, 20, 25, and 30 data points. The results demonstrated that the LyE graph reached a plateau at the 15-point neighboring condition implying that the LyE values calculated using at least 15 neighboring points were consistent. The proposed method could be used to calculate more consistent LyE values in experimental time series acquired from biological systems where noise is omnipresent.