A Smartphone-Based Tool for Assessing Parkinsonian Hand Tremor

Area of Science:

• Neurology
• Biomedical Engineering
• Digital Health

Background:

• Parkinson's disease (PD) diagnosis and monitoring often rely on clinical assessments that can be subjective.
• Objective quantification of upper limb tremor in PD patients is crucial for effective management and research.

Purpose of the Study:

• To develop and validate a practical, smartphone-based tool for accurate assessment of upper limb tremor in Parkinson's disease patients.
• To explore the utility of machine learning algorithms in classifying PD patients based on sensor-derived tremor metrics.

Main Methods:

• Utilized smartphone accelerometer and gyroscope signals to compute tremor-related metrics.
• Collected data from 25 PD patients and 20 age-matched healthy volunteers.
• Applied machine learning techniques to classify subjects based on the computed metrics.

Main Results:

• The combined metrics and machine learning approach achieved 82% accuracy in classifying PD patients.
• Achieved 90% accuracy in classifying healthy volunteers, demonstrating high diagnostic potential.
• The tool is low-cost, platform-independent, non-invasive, and user-friendly.

Conclusions:

• The smartphone-based tool offers a practical and accurate method for assessing upper limb tremor in Parkinson's disease.
• This technology can support clinical decision-making, facilitate remote patient monitoring, and aid in PD research.
• The system's accessibility and ease of use can enhance patient-physician connectivity and data collection for longitudinal studies.