Dataset Information

Automated Movement Assessment in Stroke Rehabilitation.

ABSTRACT: We are developing a system for long term Semi-Automated Rehabilitation At the Home (SARAH) that relies on low-cost and unobtrusive video-based sensing. We present a cyber-human methodology used by the SARAH system for automated assessment of upper extremity stroke rehabilitation at the home. We propose a hierarchical model for automatically segmenting stroke survivor's movements and generating training task performance assessment scores during rehabilitation. The hierarchical model fuses expert therapist knowledge-based approaches with data-driven techniques. The expert knowledge is more observable in the higher layers of the hierarchy (task and segment) and therefore more accessible to algorithms incorporating high level constraints relating to activity structure (i.e., type and order of segments per task). We utilize an HMM and a Decision Tree model to connect these high level priors to data driven analysis. The lower layers (RGB images and raw kinematics) need to be addressed primarily through data driven techniques. We use a transformer based architecture operating on low-level action features (tracking of individual body joints and objects) and a Multi-Stage Temporal Convolutional Network(MS-TCN) operating on raw RGB images. We develop a sequence combining these complimentary algorithms effectively, thus encoding the information from different layers of the movement hierarchy. Through this combination, we produce a robust segmentation and task assessment results on noisy, variable and limited data, which is characteristic of low cost video capture of rehabilitation at the home. Our proposed approach achieves 85% accuracy in per-frame labeling, 99% accuracy in segment classification and 93% accuracy in task completion assessment. Although the methodology proposed in this paper applies to upper extremity rehabilitation using the SARAH system, it can potentially be used, with minor alterations, to assist automation in many other movement rehabilitation contexts (i.e., lower extremity training for neurological accidents).

SUBMITTER: Ahmed T

PROVIDER: S-EPMC8417323 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundAs many as 80% of stroke survivors experience upper limb (UL) disability. The strong relationships between disability, lost productivity, and ongoing health care costs mean reducing disability after stroke is critical at both individual and society levels. Unfortunately, the amount of UL-focused rehabilitation received by people with stroke is extremely low. Activity monitoring and promotion using wearable devices offer a potential technology-based solution to address this gap. Commonly, wearable devices are used to deliver a haptic nudge to the wearer with the aim of promoting a particular behavior. However, little is known about the effectiveness of haptic nudging in promoting behaviors in patient populations.ObjectiveThis study aimed to estimate the effect of haptic nudging delivered via a wrist-worn wearable device on UL movement in people with UL disability following stroke undertaking inpatient rehabilitation.MethodsA multiple-period randomized crossover design was used to measure the association of UL movement with the occurrence of haptic nudge reminders to move the affected UL in 20 people with stroke undertaking inpatient rehabilitation. UL movement was observed and classified using movement taxonomy across 72 one-minute observation periods from 7:00 AM to 7:00 PM on a single weekday. On 36 occasions, a haptic nudge to move the affected UL was provided just before the observation period. On the other 36 occasions, no haptic nudge was given. The timing of the haptic nudge was randomized across the observation period for each participant. Statistical analysis was performed using mixed logistic regression. The effect of a haptic nudge was evaluated from the intention-to-treat dataset as the ratio of the odds of affected UL movement during the observation period following a "Planned Nudge" to the odds of affected limb movement during the observation period following "No Nudge."ResultsThe primary intention-to-treat analysis showed the odds ratio (OR) of affected UL movement following a haptic nudge was 1.44 (95% CI 1.28-1.63, P<.001). The secondary analysis revealed an increased odds of affected UL movement following a Planned Nudge was predominantly due to increased odds of spontaneous affected UL movement (OR 2.03, 95% CI 1.65-2.51, P<.001) rather than affected UL movement in conjunction with unaffected UL movement (OR 1.13, 95% CI 0.99-1.29, P=.07).ConclusionsHaptic nudging delivered via a wrist-worn wearable device increases affected UL movement in people with UL disability following stroke undertaking inpatient rehabilitation. The promoted movement appears to be specific to the instructions given.Trial registrationAustralia New Zealand Clinical Trials Registry 12616000654459; https://www.anzctr.org.au/Trial/Registration/TrialReview.aspx?id=370687&isReview=true.

Project description:ObjectivesThis study aimed to describe the frequency, determinants and outcomes for assessment of patients who had an acute ischaemic stroke (AIS) for rehabilitation during hospitalisation in China.DesignA registry-based retrospective observational study.Study design and settingsData regarding assessment or rehabilitation were extracted from the Chinese Stroke Center Alliance database from 1 August 2015 to 31 July 2019. Univariate and multivariate analyses were conducted to identify patient and hospital characteristics associated with rehabilitation assessment during acute hospitalisation as well as discharge outcomes.Study cohortWe included 837 897 patients who had a stroke in this study with patient characteristics, admission location, medical history, hospital characteristics and hospital designation.Primary and secondary outcome measuresRehabilitation assessment and discharge outcomes.ResultsAmong 837 897 patients who had a stroke admitted to 1473 hospitals, 615 991 (73.5%) underwent rehabilitation assessment. There were significant variations in the rates of rehabilitation assessment across hospitals (IQR 61.3% vs 92.9%). According to multivariate analysis, guideline recommended care delivery was associated with a higher rehabilitation assessment rate, whereas high/low body mass index, ambulation (OR 0.88; 95% CI 0.87 to 0.90), history of stroke (OR 0.94; 95% CI 0.93 to 0.95), coronary heart disease (OR 0.84; 95% CI 0.82 to 0.85) and atrial fibrillation (OR 0.91; 95% CI 0.89 to 0.94) were associated with a lower rate. Additionally, rehabilitation assessment during hospitalisation was significantly associated with lower in-hospital mortality (OR 0.38; 95% CI 0.35 to 0.41) and a higher probability of discharge to a rehabilitation centre (OR 2.66; 95% CI 2.5 to 2.82).ConclusionsNearly one-quarter of patients who had an AIS do not undergo documented rehabilitation assessment and compliance across hospitals varies. Thus, it is necessary to improve adherence to rehabilitation assessment to improve the quality of medical care for patients who had an AIS.

Dataset Information

Automated Movement Assessment in Stroke Rehabilitation.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets