Non-Contact Early Warning of Shaking Palsy.

Area of Science:

• Biomedical Engineering
• Neurology
• Signal Processing

Background:

• Parkinsonian gait significantly impacts quality of life in shaking palsy (SP).
• Current diagnostic methods may not fully capture gait abnormalities.
• Objective gait analysis is crucial for early SP detection and management.

Purpose of the Study:

• To develop and validate a non-contact system for detecting Parkinsonian gait.
• To classify abnormal gait in SP patients versus normal walking using S-band signals.
• To enable early warning of SP through gait analysis.

Main Methods:

• Utilized S-band perception technique for gait analysis.
• Implemented data preprocessing for amplitude and phase information.
• Applied Principal Component Analysis (PCA) for feature extraction.
• Employed Support Vector Machine (SVM) for gait classification.

Main Results:

• The proposed system achieved high accuracy (above 90%) in detecting Parkinsonian gait.
• Demonstrated successful classification of abnormal gait in SP patients.
• Validated the system's effectiveness across diverse participants.

Conclusions:

• The S-band perception technique offers a promising non-contact method for SP diagnosis.
• High accuracy supports the system's potential for early SP detection.
• This approach can improve patient monitoring and management of shaking palsy.