A Low-Cost Wearable System to Support Upper Limb Rehabilitation in Resource-Constrained Settings

Abstract

There is a lack of professional rehabilitation therapists and facilities in low-resource settings such as Bangladesh. In particular, the restrictively high costs of rehabilitative therapy have prompted a search for alternatives to traditional in-patient/out-patient hospital rehabilitation moving therapy outside healthcare settings. Considering the potential for home-based rehabilitation, we implemented a low-cost wearable system for 5 basic exercises of upper limb (UL) rehabilitation through incorporation of physiotherapists’ perspectives. As a proof of concept, we collected data through our system from 10 Bangladeshi participants. Leveraging the system’s sensed data, we developed a diverse set of machine learning models. Also, we selected important features through 3 feature selection approaches. We find that the Multilayer Perceptron model, which was developed by Random Forest selected features, can identify the 5 exercises with a ROC-AUC score of 98.2% and sensitivity of 98%. Our system has the potential for getting real-time insights regarding the precision of the exercises which can facilitate home-based UL rehabilitation in resource-constrained settings.

Note:
Funding Information: This study was supported by seed funding from Cardiff University’s GCRF QR Funding from the Higher Education Funding Council for Wales.

Conflict of Interests: We do not have any competing interests.

Ethical Approval: The study was approved by the North South University IRB/ERC committee (2020/OR-NSU/IRB-No.0501). Signed consent forms were obtained from the partici- pants and also their data were anonymized.

Keywords: Upper limb rehabilitation, Low-resource, Wearable, Machine learning, Exercises, Physiotherapy, Bangladesh