Cardiff University | Prifysgol Caerdydd ORCA
Online Research @ Cardiff 
WelshClear Cookie - decide language by browser settings

Eldo-care: EEG with Kinect sensor based telehealthcare for the disabled and the elderly

Das, Sima, Adhikary, Arpan, Laghari, Asif Ali and Mitra, Solanki 2023. Eldo-care: EEG with Kinect sensor based telehealthcare for the disabled and the elderly. Neuroscience Informatics 3 (2) , 100130. 10.1016/j.neuri.2023.100130

[thumbnail of eldocare.pdf]
PDF - Published Version
Available under License Creative Commons Attribution Non-commercial No Derivatives.

Download (1MB) | Preview


Telehealthcare systems are nowadays becoming a massive daily helping kit for elderly and disabled people. By using the Kinect sensors, remote monitoring has become easy. Also, the sensors' data are useful for the further improvement of the device. In this paper, we have discussed our newly developed “Eldo-care” system. This system is designed for the assessment and management of diverse neurological illnesses. The telemedical system is developed to monitor the psycho-neurological condition. People with disabilities and the elderly frequently experience access issues to essential services. Researchers today are concentrating on rehabilitative technologies based on human-computer interfaces that are closer to social-emotional intelligence. The goal of the study is to help old and disabled persons with cognitive rehabilitation using machine learning techniques. Human brain activity is observed using electroencephalograms, while user movement is tracked using Kinect sensors. Chebyshev filter is used for feature extraction and noise reduction. Utilizing the autoencoder technique, categorization is carried out by a Convolutional neural network with an accuracy of 95% and higher based on transfer learning. A better quality of life for older and disabled persons will be attained through the application of the suggested system in real time. The proposed device is attached to the subject under monitoring.

Item Type: Article
Date Type: Publication
Status: Published
Schools: Engineering
ISSN: 2772-5286
Date of First Compliant Deposit: 17 January 2024
Date of Acceptance: 27 April 2023
Last Modified: 19 Jan 2024 17:00

Actions (repository staff only)

Edit Item Edit Item


Downloads per month over past year

View more statistics