Luckshman Bavan1, Karl Surmacz2, David Beard3, Stephen Mellon4, Jonathan Rees5. 1. Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, University of Oxford, Botnar Research Centre, Old Road, Oxford, OX3 7LD, United Kingdom. Electronic address: Luckshman.Bavan@ndorms.ox.ac.uk. 2. McLaren Applied Technologies, McLaren Technology Centre, Chertsey Road, Woking, GU21 4YH, United Kingdom. Electronic address: Karl.Surmacz@mclaren.com. 3. Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, University of Oxford, Botnar Research Centre, Old Road, Oxford, OX3 7LD, United Kingdom. Electronic address: David.Beard@ndorms.ox.ac.uk. 4. Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, University of Oxford, Botnar Research Centre, Old Road, Oxford, OX3 7LD, United Kingdom. Electronic address: Stephen.Mellon@ndorms.ox.ac.uk. 5. Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, University of Oxford, Botnar Research Centre, Old Road, Oxford, OX3 7LD, United Kingdom. Electronic address: Jonathan.Rees@ndorms.ox.ac.uk.
Abstract
BACKGROUND: 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. RESEARCH QUESTION: 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. METHODS: 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. RESULTS: 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. SIGNIFICANCE: 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.
BACKGROUND: 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. RESEARCH QUESTION: 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. METHODS: 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. RESULTS: 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. SIGNIFICANCE: 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.
Authors: Up Huh; Young Jin Tak; Seunghwan Song; Sung Woon Chung; Sang Min Sung; Chung Won Lee; Miju Bae; Hyo Young Ahn Journal: JMIR Mhealth Uhealth Date: 2019-06-18 Impact factor: 4.773