A Comprehensive Multifunctional Approach for Measuring Parkinson's Disease Severity

Area of Science:

• Neuroscience
• Artificial Intelligence
• Medical Diagnostics

Background:

• Current Parkinson's disease (PD) staging predominantly relies on motor symptoms.
• Existing staging systems may not fully capture the multifaceted nature of PD progression.

Purpose of the Study:

• To develop an advanced PD staging framework using machine learning.
• To integrate a wider range of neurocognitive symptoms into PD staging.
• To create a more objective and personalized staging system.

Main Methods:

• Recruited 37 individuals diagnosed with PD.
• Administered tablet-based neurocognitive tests covering motor, memory, speech, and executive functions.
• Developed a hybrid feature scoring system using random forest and principal component analysis.

Main Results:

• Current PD staging shows a bias towards fine motor skills.
• Neurocognitive functions like memory, speech, and executive function are underrepresented in current staging.
• A more comprehensive assessment requires including a broader spectrum of neurocognitive functions.

Conclusions:

• The proposed hybrid feature score offers a more holistic understanding of PD.
• This approach can lead to more effective, objective, and personalized treatment strategies.
• The methodology is adaptable for staging other neurodegenerative diseases.