EventsThe 1st International Online Conference on Non-Destructive Testing
Published
This submission belongs to the session S3. Advanced Sensing Technologies of the event The 1st International Online Conference on Non-Destructive Testing
Published date
26 Jun, 2026
Academic Editor
author-avatarFabio Tosti
Citation
Ahmad Naim Zarwi, Adnan Abdulkarim Alsaei, Ayah Adel Binrajab, Fatema Nader Rahimi, G. Roshan Deen, AI Wearable Biosensing Device Equipped with Inertial Measurement Sensors and Electromyography Sensors for Early Detection of Parkinson’s Disease, in Proceedings of The 1st International Online Conference on Non-Destructive Testing, 1 July–3 July 2026, MDPI: Basel, Switzerland
Share
Email
Facebook
Twitter
LinkedIn

AI Wearable Biosensing Device Equipped with Inertial Measurement Sensors and Electromyography Sensors for Early Detection of Parkinson’s Disease

Adnan Abdulkarim Alsaei 1
Ayah Adel Binrajab 1
Fatema Nader Rahimi 1
image
1. School of Medicine, Royal College of Surgeons in Ireland - Bahrain (RCSI), Busaiteen , Kingdom of Bahrain, Bahrain
Abstract

Parkinson’s disease is a progressive neurodegenerative disease resulting in motor symptoms including muscle rigidity, resting tremor, bradykinesia, and postural instability. Given that the diagnosis is primarily clinical, early symptoms can be overlooked, underscoring the need for improved methods of early detection. Wearable biosensor devices that use highly sensitive motion sensors can detect slight motor fluctuations including subtle tremors and gait irregularities. Data collected from these devices can be analyzed by machine learning algorithms to identify motor abnormalities associated with Parkinson’s disease, enabling early diagnosis and improved management. The aim of this study is to develop a wearable biosensor device integrated with machine learning to detect subtle early motor changes that may be linked to Parkinson’s disease. The wearable device is designed to integrate inertial measurement unit sensors (IMU) including accelerometers, which detect tremor amplitude; gyroscopes, which measure angular velocity to determine tremor frequency; and magnetometers, which track posture and rotation. The device also incorporates electromyography (EMG) sensors to monitor rigidity, arm swing, and muscle contraction patterns, in addition to physiological sensors to track autonomic function. Machine learning training will be conducted to allow pattern identification of motor changes consistent with Parkinsonian motor dysfunction. Validity will be assessed by comparing sensor-derived measurements with clinical evaluations to determine its detection reliability. We expect that wearing these devices will enable accurate detection of early Parkinson’s disease symptoms allowing for early treatment intervention and potentially slowing disease progression. Early detection of Parkinson’s disease identifies pre-motor and non-motor signs, occurring before the loss of over 50% of dopaminergic neurons, facilitating timely neuroprotective treatment that may slow symptom progression.

Keywords
Biosensors
Parkinson's Disease
Machine Learning
Wearable Device
AI
Inertial Measurement Unit Sensors
Examination of filled polymers by AFM phase imaging
Defect Evaluation in Ferromagnetic Materials Using a Custom-Designed Instrument for Magnetic Barkhausen Noise and Hysteresis Loop Measurements