Using fuzzy logic for diagnosis and classification of spasticity

Area of Science:

• Biomedical Engineering
• Neuroscience
• Signal Processing

Background:

• Spasticity is a complex sensory-motor disorder lacking clear pathophysiological mechanisms and standardized evaluation methods.
• Existing diagnostic approaches face challenges in quantitative measurement and comprehensive assessment.

Purpose of the Study:

• To develop a quantitative diagnosis system for spasticity using a fuzzy logic classifier (FLC).
• To investigate the underlying pathophysiological mechanisms of spasticity through surface electromyography (EMG) analysis.

Main Methods:

• Surface EMG signals were recorded from tibialis anterior and medial gastrocnemius muscles in hemiplegic patients and healthy controls.
• EMG signals were analyzed using time-domain amplitudes, short-time Fourier transform, and wavelet transform.
• A Mamdani-type fuzzy logic classifier with 64 rules was developed for signal analysis.

Main Results:

• Wavelet transform demonstrated superior performance in analyzing EMG signals compared to other methods.
• The FLC system achieved 100% sensitivity, 95.8% accuracy, and 93.8% specificity in identifying spasticity.

Conclusions:

• A fuzzy logic classifier system was successfully developed for detecting and identifying spasticity.
• The FLC system demonstrates high diagnostic performance despite inherent measurement challenges in spasticity.