Parkinson's disease hand tremor detection system for mobile application.

Area of Science:

• Biomedical Engineering
• Neurology
• Mobile Health (mHealth)

Background:

• Parkinson's disease (PD) is a progressive neurodegenerative disorder affecting millions globally.
• Current PD management focuses on symptom treatment, necessitating accurate symptom quantification, particularly hand tremor.
• No cure exists for PD; early detection and monitoring are crucial for effective management.

Purpose of the Study:

• To develop and validate a novel system for detecting Parkinson's disease rest tremor using smartphone accelerometers.
• To assess the feasibility of utilizing mobile phone technology for objective PD symptom measurement.
• To differentiate between Parkinson's disease patients and healthy individuals based on tremor patterns.

Main Methods:

• Utilized smartphone accelerometers to record hand movements in 21 PD patients and 21 healthy controls.
• Applied a two-level wavelet packet analysis to extract 12 distinct features from the accelerometer data.
• Classified the extracted features using a neural network classifier to distinguish between PD patients and controls.

Main Results:

• The developed system achieved a high accuracy of 95% in detecting Parkinson's disease rest tremor.
• A Kappa coefficient of 90% was obtained, indicating strong agreement and reliability of the classification.
• The study successfully differentiated PD patients from healthy subjects based on accelerometer-derived tremor analysis.

Conclusions:

• Smartphone accelerometers offer a non-invasive, accessible, and accurate method for detecting and quantifying Parkinson's disease rest tremor.
• This mHealth approach holds significant potential for remote patient monitoring and aiding in PD diagnosis.
• Further research can explore integration into routine clinical practice for improved PD management.