Barbareschi, G; Holloway, CSM; Berthouze, N; Sonenblum, S; Sprigle, S; (2018) Use of a Low Cost, Chest-Mounted Accelerometer to Evaluate Transfer Skills of Wheelchair Users During Everyday Activities. JMIR Rehabilitation and Assistive Technologies , 5 (2) , Article e11748. 10.2196/11748.

Preview

Text Barbareschi_Use of a low-cost, chest-mounted accelerometer to evaluate transfer skills of wheelchair users during.pdf - Published Version Download (1MB) | Preview

Abstract

BACKGROUND: Transfers are an important skill for many wheelchair users. However, they have also been related to the risk of falling or developing upper limb injuries. Transfer abilities are usually evaluated in clinical settings or biomechanics laboratories and these methods of assessment are poorly suited to evaluation in real and unconstrained world settings where transfers take place. OBJECTIVE: The objective of this paper is to develop a strategy to enable transfer quality evaluation and improve the predictive accuracy of transfer detection using a single wearable low cost accelerometer. METHODS: We collected data from nine wheelchair users wearing tri-axial accelerometer on their chest while performing transfers to and from car seats and home furniture. We then extracted significant features from accelerometer data based on biomechanical considerations and previous relevant literature and used machine learning algorithms to evaluate the performance of wheelchair transfers and detect their occurrence from a continuous time series of data. RESULTS: Results show that the best predictive accuracy for Automatic Transfer Quality Evaluation was obtained with Support Vector Machine (SVM) classifiers when determining use of head-hip relationship (75.93%) and smoothness of landing (79.62%), when the start and end of the transfer are known. Automatic Transfer Detection reaches an accuracy of 87.8% using Multinomial Logistic Regression (MLR) classifiers, which is in line with the state of the art in this context. However, we achieve these results using only a single sensor and collecting data in a more ecological manner. CONCLUSIONS: The use of a single chest-placed accelerometer shows a predictive accuracy of over 75% for algorithms applied independently to both transfer evaluation and monitoring. This points to the opportunity for designing ubiquitous technology for personalized skill development interventions targeting wheelchair users. However, monitoring transfers still requires the use of external inputs or extra sensors to identify start and end of the transfer, which are needed to perform an accurate evaluation.

Download activity - last month

Export as CSVXMLJSON

Download activity - last 12 months

Export as JSONXMLCSV

Downloads by country - last 12 months

Export as JSONCSVXML

Archive Staff Only

View Item