Open Access

ARTICLE

IoMT-Based Healthcare Framework for Ambient Assisted Living Using a Convolutional Neural Network

Waleed T. Al-Sit¹, Nidal A. Al-Dmour², Taher M. Ghazal^3,4,*, Ghassan F. Issa³

1 Department of Computer Engineering, Mutah University, Al-Karak, Jordan; Higher Colleges of Technology, Dubai, UAE
2 Department of Computer Engineering, College of Engineering, Mutah University, Jordan
3 School of Information Technology, Skyline University College, University City Sharjah, Sharjah, UAE
4 Center for Cyber Security, Faculty of Information Science and Technology, UKM, 43600, Bangi, Selangor, Malaysia

* Corresponding Author: Taher M. Ghazal. Email:

Computers, Materials & Continua 2023, 74(3), 6867-6878. https://doi.org/10.32604/cmc.2023.034952

Received 02 August 2022; Accepted 01 November 2022; Issue published 28 December 2022

Abstract

In the age of universal computing, human life is becoming smarter owing to the recent developments in the Internet of Medical Things (IoMT), wearable sensors, and telecommunication innovations, which provide more effective and smarter healthcare facilities. IoMT has the potential to shape the future of clinical research in the healthcare sector. Wearable sensors, patients, healthcare providers, and caregivers can connect through an IoMT network using software, information, and communication technology. Ambient assisted living (AAL) allows the incorporation of emerging innovations into the routine life events of patients. Machine learning (ML) teaches machines to learn from human experiences and to use computer algorithms to “learn” information directly instead of relying on a model. As the sample size accessible for learning increases, the performance of the algorithms improves. This paper proposes a novel IoMT-enabled smart healthcare framework for AAL to monitor the physical actions of patients using a convolutional neural network (CNN) algorithm for fast analysis, improved decision-making, and enhanced treatment support. The simulation results showed that the prediction accuracy of the proposed framework is higher than those of previously published approaches.

---

Keywords

---