Classification and visualization tool for gait analysis of Parkinson's disease

Area of Science:

• Neurology
• Biostatistics
• Data Mining

Background:

• Parkinson's Disease (PD) lacks a definitive diagnostic test.
• Current diagnostic methods rely on clinical symptoms and patient history.
• There is a need for objective and accurate diagnostic tools.

Purpose of the Study:

• To develop a data mining-based statistical method for diagnosing Parkinson's Disease.
• To create a standard and accurate diagnostic test for PD.
• To evaluate the accuracy of the proposed diagnostic method.

Main Methods:

• Utilized data mining techniques to analyze patient data.
• Developed a statistical regression formula based on stride data.
• Validated the diagnostic accuracy of the proposed method.

Main Results:

• The proposed regression formula achieved 90% accuracy in diagnosing Parkinson's Disease.
• The method requires only patient stride data for diagnosis.
• Demonstrated the potential for a standardized and accurate PD diagnostic test.

Conclusions:

• A data mining-based statistical approach can accurately diagnose Parkinson's Disease.
• Patient stride data is a viable biomarker for PD diagnosis.
• The proposed method offers a promising avenue for a standard, accurate PD diagnostic test.