Decision support framework for Parkinson's disease based on novel handwriting markers

Area of Science:

• Neurology
• Biomedical Engineering
• Data Science

Background:

• Parkinson's disease (PD) is a neurodegenerative disorder affecting motor and non-motor functions.
• Handwriting deterioration is a common early clinical hallmark of PD.

Purpose of the Study:

• Identify specific handwriting features for PD detection.
• Develop a predictive model for efficient PD diagnosis.

Main Methods:

• Collected handwriting samples from 37 PD patients and 38 controls.
• Extracted conventional and novel handwriting measures (kinematic, spatio-temporal, entropy, signal energy, empirical mode decomposition).
• Utilized a support vector machine classifier with a radial Gaussian kernel for automated diagnosis.

Main Results:

• Achieved 88.13% accuracy in classifying PD.
• Demonstrated high sensitivity (89.47%) and specificity (91.89%) in PD detection.
• Identified a subset of handwriting features effective for PD diagnosis.

Conclusions:

• Handwriting analysis shows potential as a valuable tool for PD diagnosis and screening.
• Novel handwriting measures contribute to improved PD detection accuracy.
• Automated analysis of handwriting offers a promising approach for early Parkinson's disease identification.