Tolerance-based evaluation of a Shariah-compliant rehabilitation IoT framework for post-stroke hand gesture monitoring

Zainuddin, Ahmad Anwar and Islam, Md Sariful and Handayani, Dini Oktarina Dwi and Kamarudin, Saidatul Izyanie and Mohd. Tamrin, Mohd. Izzuddin and Misman, Ahmad Fatzilah (2026) Tolerance-based evaluation of a Shariah-compliant rehabilitation IoT framework for post-stroke hand gesture monitoring. In: 2025 8th International Conference on Information and Communications Technology (ICOIACT), 4-5 December 2025, YOGYAKARTA.

PDF - Published Version Restricted to Registered users only Download (504kB) | Request a copy

Abstract

The lack of affordable, continuous, and inclusive rehabilitation services creates disparities in patient outcomes, particularly for vulnerable groups, and raises concerns about fulfilling the ethical responsibility to ensure fair and compassionate healthcare delivery. To address these challenges within a Shariah-compliant healthcare framework, this study introduces a computational approach using the Rehabilitation Internet of Things (RIoT) to enable remote, ethical, and Patientcentred rehabilitation. Hand Gesture-based data from 200 poststroke patients at Putrajaya Hospital were collected using the MediaPipe Pose framework for real-time skeletal and hand tracking. Features such as repetition count, and completion time were extracted and analysed through predictive models, including linear regression and ensemble methods, with validation based on tolerance-based accuracy. Tolerance-based accuracy evaluation demonstrated clinically meaningful outcomes within a 20%–30% margin of agreement with manual assessments. For hand strengthening (HS), the system achieved 71.5% accuracy at ±20% tolerance and 88.5% at ±30%, reflecting strong reliability in measuring gross motor functions. Hand opposition (HO), which relies on fine motor precision, yielded 61.5% accuracy at ±20% and 84.5% at ±30%, indicating acceptable reliability at broader thresholds despite higher variability. These results affirm that RIoT can be regarded as a clinically usable tool for remote rehabilitation monitoring, particularly under ±30% tolerance. Furthermore, significant correlations were observed between computational metrics and established clinical outcomes, confirming the reliability of the proposed framework. Beyond clinical utility, the framework ensures patient data privacy and aligns with the Maqasid alShariah in upholding ʿAdl (justice), Taysīr (facilitation of access), and preservation of Akhlāq (morality). Overall, the findings highlight the potential for ASEAN healthcare systems to adopt technology-driven rehabilitation strategies that enhance patient autonomy.

Item Type:	Proceeding Paper (Other)
Uncontrolled Keywords:	Rehabilitation Internet of Things (RIoT), poststroke rehabilitation, Computational modelling, Shariahcompliant -healthcare, Gesture recognition.
Subjects:	T Technology > T Technology (General) > T10.5 Communication of technical information
Kulliyyahs/Centres/Divisions/Institutes (Can select more than one option. Press CONTROL button):	Kulliyyah of Information and Communication Technology > Department of Computer Science Kulliyyah of Information and Communication Technology > Department of Computer Science Kulliyyah of Information and Communication Technology Kulliyyah of Information and Communication Technology
Depositing User:	Ts.Dr. Ahmad Anwar Zainuddin
Date Deposited:	19 Feb 2026 13:10
Last Modified:	19 Feb 2026 13:10
Queue Number:	2026-02-Q2146
URI:	http://irep.iium.edu.my/id/eprint/127432

Actions (login required)

View Item