Gait is a core motor function and is impaired in numerous neurological diseases, including Parkinson's disease (PD). Treatment changes in PD are frequently driven by gait assessments in the clinic, ...commonly rated as part of the Movement Disorder Society (MDS) Unified PD Rating Scale (UPDRS) assessment (item 3.10). We proposed and evaluated a novel approach for estimating severity of gait impairment in Parkinson's disease using a computer vision-based methodology. The system we developed can be used to obtain an estimate for a rating to catch potential errors, or to gain an initial rating in the absence of a trained clinician-for example, during remote home assessments. Videos (n=729) were collected as part of routine MDS-UPDRS gait assessments of Parkinson's patients, and a deep learning library was used to extract body key-point coordinates for each frame. Data were recorded at five clinical sites using commercially available mobile phones or tablets, and had an associated severity rating from a trained clinician. Six features were calculated from time-series signals of the extracted key-points. These features characterized key aspects of the movement including speed (step frequency, estimated using a novel Gamma-Poisson Bayesian model), arm swing, postural control and smoothness (or roughness) of movement. An ordinal random forest classification model (with one class for each of the possible ratings) was trained and evaluated using 10-fold cross validation. Step frequency point estimates from the Bayesian model were highly correlated with manually labelled step frequencies of 606 video clips showing patients walking towards or away from the camera (Pearson's r=0.80, p<0.001). Our classifier achieved a balanced accuracy of 50% (chance = 25%). Estimated UPDRS ratings were within one of the clinicians' ratings in 95% of cases. There was a significant correlation between clinician labels and model estimates (Spearman's ρ=0.52, p<0.001). We show how the interpretability of the feature values could be used by clinicians to support their decision-making and provide insight into the model's objective UPDRS rating estimation. The severity of gait impairment in Parkinson's disease can be estimated using a single patient video, recorded using a consumer mobile device and within standard clinical settings; i.e., videos were recorded in various hospital hallways and offices rather than gait laboratories. This approach can support clinicians during routine assessments by providing an objective rating (or second opinion), and has the potential to be used for remote home assessments, which would allow for more frequent monitoring.
Background. The Movement Disorder Society Unified Parkinson’s Disease Rating Scale (MDS-UPDRS) comprises 50 items, consisting of historical questions and motor ratings, typically taking around 30 ...minutes to complete. We sought to identify an abbreviated version that could facilitate use in clinical practice or used remotely via telemedicine. Methods. To create an 8-item version we conducted an “exhaustive search” of all possible subsets. We measured explained variance in comparison to the 50-item version using linear regression, with the “optimal” subset maximising this while also meeting remote assessment practicality constraints. The subset was identified using a dataset collected by the Parkinson’s Progression Markers Initiative and validated using an MDS Non-Motor Symptoms Scale validation study dataset. Results. The optimal remote version comprised items from all parts of the MDS-UPDRS and was found to act as an unbiased estimator of the total 50-item score. This version had an explained variance score of 0.844 and was highly correlated with the total MDS-UPDRS score (Pearson’s r = 0.919, p-value <0.0001). Another subset that maximised explained variance score without adhering to remote assessment practicality constraints provided similar results. Conclusion. This result demonstrates that the total scores of an abbreviated form identified by computational statistics had high agreement with the MDS-UPDRS total score. Whilst it cannot capture the richness of information of the full MDS-UPDRS, it can be used to create a total score where practicality limits the application of the full MDS-UPDRS, such as remote monitoring. Further validation will be required, including in specific subgroups and advanced disease stages, and full validation of clinimetric properties.
Parkinson's disease (PD) is a common neurological disorder, with bradykinesia being one of its cardinal features. Objective quantification of bradykinesia using computer vision has the potential to ...standardise decision-making, for patient treatment and clinical trials, while facilitating remote assessment. We utilised a dataset of part-3 MDS-UPDRS motor assessments, collected at four independent clinical and one research sites on two continents, to build computer-vision-based models capable of inferring the correct severity rating robustly and consistently across all identifiable subgroups of patients. These results contrast with previous work limited by small sample sizes and small numbers of sites. Our bradykinesia estimation corresponded well with clinician ratings (interclass correlation 0.74). This agreement was consistent across four clinical sites. This result demonstrates how such technology can be successfully deployed into existing clinical workflows, with consumer-grade smartphone or tablet devices, adding minimal equipment cost and time.
Over the last decade, video-enabled mobile devices have become ubiquitous, while advances in markerless pose estimation allow an individual's body position to be tracked accurately and efficiently ...across the frames of a video. Previous work by this and other groups has shown that pose-extracted kinematic features can be used to reliably measure motor impairment in Parkinson's disease (PD). This presents the prospect of developing an asynchronous and scalable, video-based assessment of motor dysfunction. Crucial to this endeavour is the ability to automatically recognise the class of an action being performed, without which manual labelling is required. Representing the evolution of body joint locations as a spatio-temporal graph, we implement a deep-learning model for video and frame-level classification of activities performed according to part 3 of the Movement Disorder Society Unified PD Rating Scale (MDS-UPDRS). We train and validate this system using a dataset of n=7310 video clips, recorded at 5 independent sites. This approach reaches human-level performance in detecting and classifying periods of activity within monocular video clips. Our framework could support clinical workflows and patient care at scale through applications such as quality monitoring of clinical data collection, automated labelling of video streams, or a module within a remote self-assessment system.
The ability to arise from a sitting to a standing position is often impaired in Parkinson's disease (PD). This impairment is associated with an increased risk of falling, and higher risk of dementia. ...We propose a novel approach to estimate Movement Disorder Society Unified PD Rating Scale (MDS-UPDRS) ratings for “item 3.9” (arising from chair) using a computer vision-based method, whereby we use clinically informed reasoning to engineer a small number of informative features from high dimensional markerless pose estimation data.
We analysed 447 videos collected via the KELVIN-PD™ platform, recorded in clinical settings at multiple sites, using commercially available mobile smart devices. Each video showed an examination for item 3.9 of the MDS-UPDRS and had an associated severity rating from a trained clinician on the 5-point scale (0, 1, 2, 3 or 4).
The deep learning library OpenPose was used to extract pose estimation key points from each frame of the videos, resulting in time-series signals for each key point. From these signals, features were extracted which capture relevant characteristics of the movement; velocity variation, smoothness, whether the patient used their hands to push themselves up, how stooped the patient was while sitting and how upright the patient was when fully standing. These features were used to train an ordinal classification system (with one class for each of the possible ratings on the UPDRS), based on a series of random forest classifiers.
The UPDRS ratings estimated by this system, using leave-one-out cross validation, corresponded exactly to the ratings made by clinicians in 79% of videos, and were within one of those made by clinicians in 100% of cases. The system was able to distinguish normal from Parkinsonian movement with a sensitivity of 62.8% and a specificity of 90.3%. Analysis of misclassified examples highlighted the potential of the system to detect potentially mislabelled data.
We show that our computer-vision based method can accurately quantify PD patients’ ability to perform the arising from chair action. As far as we are aware this is the first study estimating scores for item 3.9 of the MDS-UPDRS from singular monocular video. This approach can help prevent human error by identifying unusual clinician ratings, and provides promise for such a system being used routinely for clinical assessments, either locally or remotely, with potential for use as stratification and outcome measures in clinical trials.
•The ability to arise from a chair, often impaired in Parkinson’s disease, is commonly assessed subjectively by clinicians.•A novel approach, using 2D pose estimation and machine learning, can robustly estimate disease severity using only videos.•The system is a good candidate for routine in office use or for remote home assessment of patients.