ORCID Profile
0000-0002-6375-6839
Current Organisation
University of Oxford
Does something not look right? The information on this page has been harvested from data sources that may not be up to date. We continue to work with information providers to improve coverage and quality. To report an issue, use the Feedback Form.
Publisher: Elsevier BV
Date: 2021
Publisher: Elsevier BV
Date: 05-2019
DOI: 10.1016/J.GAITPOST.2019.03.008
Abstract: Rehabilitation has an established role in the management of a wide range of musculoskeletal conditions. Much of this treatment relies on self-directed exercises at home, where adherence of execution is unknown. Demonstrating treatment fidelity is necessary to draw conclusions about the efficacy of rehabilitation interventions in both clinical and research settings. There is a lack of tools and methods to achieve this. This study aims to evaluate the feasibility of using a single inertial sensor to recognise and classify shoulder rehabilitation activity using supervised machine learning techniques. Twenty patients with shoulder pain were monitored performing five rehabilitation exercises routinely prescribed for their condition. Accelerometer, gyroscope and magnetometer data were collected via a device mounted onto an arm sleeve. Non-specific motion data was included in the analysis. Time and frequency domain features were calculated from labelled data segments and ranked in terms of their predictive importance using the ReliefF algorithm. Selected features were used to train four supervised learning algorithms: decision tree, k-nearest neighbour, support vector machine and random forests. Performance of algorithms in accurately classifying exercise activity was evaluated with ten-fold cross-validation and leave-one-subject-out-validation methods. Optimal predictive accuracies for ten-fold cross-validation (97.2%) and leave-one-subject-out-validation (80.5%) were achieved by support vector machine and random forests algorithms, respectively. Time domain features derived from accelerometer, magnetometer and orientation data streams were shown to have the highest predictive value for classifying rehabilitation activity. Classification models performed well in differentiating patient exercise activity from non-specific movement and identifying specific exercise type using inertial sensor data. A clinically useful account of home rehabilitation activity will help guide treatment strategies and facilitate methods to improve patient engagement. Future work should focus on evaluating the performance of such systems in natural and unsupervised settings.
Publisher: Ovid Technologies (Wolters Kluwer Health)
Date: 06-02-2019
Abstract: There is controversy about optimal limb alignment following knee replacement. An aim of using Oxford medial unicompartmental knee replacement (UKR) implants is to accurately restore normal ligament tension in the knee, thereby restoring normal kinematics. This return to normal tension typically results in a return to prearthritic alignment, which is frequently varus. The aim of this study was to investigate the relationship between postoperative limb alignment and postoperative patient-reported outcome and implant revision rate. We used a consecutive cohort of 891 knees with cemented Oxford medial UKR implants with a mean 10-year follow-up and recorded alignment. We grouped knees according to postoperative mechanical alignment as marked varus (estimated at 10°), mild varus (estimated at 5°), neutral, and valgus. The mean Oxford Knee Score (OKS) was calculated at 5 and 10 years postoperatively. Revision risk was assessed by survival analysis and component-time incidence rates. Postoperatively, 67 (8%) of the 891 knees were in marked varus 308 (35%), in mild varus 508 (57%), in neutral and 8 (1%), in valgus. The valgus group (8 knees) was too small for further analysis. The mean OKS (and standard deviation [SD]) at 10 years postoperatively was 41.7 ± 7 for marked varus, 40.5 ± 8 for mild varus, and 39.4 ± 9 for neutral alignment (p = 0.28). At 10 years, 92%, 85%, and 76% achieved a good or excellent OKS outcome, respectively (p = 0.02). Twelve-year survival rates were 93.3% for marked varus, 93.2% for mild varus, and 93.6% for neutral alignment, respectively (p = 0.53). Revision incidence rates per 100 component-years were 0.49 (95% confidence interval [CI], 0.2 to 1.5), 0.36 (95% CI, 0.2 to 0.7), and 0.54 (95% CI, 0.4 to 0.8), respectively, and were not significantly different (p = 0.53). Marked postoperative varus mechanical alignment of an estimated 10° was present in 8%, and mild varus of about 5° was present in 35%. Increasing varus alignment was associated with an increasing percentage of good or excellent OKS outcomes, but otherwise there were no significant differences between alignment groups in patient-reported outcome or revision rate. These data support the standard operative technique for the Oxford UKR, which aims to restore ligament tension and therefore prearthritic alignment rather than neutral mechanical alignment. Therapeutic Level IV . See Instructions for Authors for a complete description of levels of evidence.
Publisher: Springer Science and Business Media LLC
Date: 21-04-2017
Publisher: IEEE
Date: 05-2013
Location: United Kingdom of Great Britain and Northern Ireland
Location: United Kingdom of Great Britain and Northern Ireland
Location: United Kingdom of Great Britain and Northern Ireland
No related grants have been discovered for Stephen Mellon.