An efficient Gait Dynamics classification method for Neurodegenerative Diseases using Brain signals

Area of Science:

• Neuroscience
• Biomedical Engineering
• Clinical Diagnostics

Background:

• Neurodegenerative diseases (NDD) like Huntington's disease (HD), Amyotrophic Lateral Sclerosis (ALS), and Parkinson's disease (PD) primarily affect brain neurons.
• Clinical diagnosis and early detection of NDD are crucial for patient outcomes, as motor impairments, including gait cycle changes, are significant symptoms.
• Characterizing gait dynamics offers a promising avenue for the early diagnosis of various NDD.

Purpose of the Study:

• To propose and validate a gait dynamics classification method for distinguishing neurodegenerative diseases from healthy controls.
• To leverage multilevel feature extraction techniques for analyzing gait signals recorded from patients with NDD and healthy subjects.
• To enhance the accuracy and timeliness of NDD diagnosis through gait analysis.

Main Methods:

• Gait signals from the left and right feet were recorded using force sensitive resistors from 16 healthy subjects, 13 ALS, 20 HD, and 15 PD patients.
• Discrete Wavelet Transform (DWT) was employed across six levels to decompose raw gait signals and extract features.
• Classification of pathological gait signals was performed using three multilevel feature extraction techniques: Detrended Fluctuation Analysis (DFA), Positive-Negative Peak Histogram Analysis (PNPHA), and Statistical Temporal parameter Analysis (STA).

Main Results:

• The proposed gait dynamics classification method successfully distinguished between individuals with neurodegenerative diseases and the healthy control group.
• Multilevel feature extraction techniques, particularly the proposed PNPHA, demonstrated effectiveness in characterizing gait abnormalities associated with NDD.
• The experimental outcomes confirmed the capability of the method in identifying pathological gait patterns.

Conclusions:

• Gait dynamics analysis is a viable approach for the early detection and classification of neurodegenerative diseases.
• The developed classification method, utilizing DWT and advanced feature extraction, shows significant potential for clinical application in NDD diagnosis.
• Accurate and timely diagnosis of NDD through gait characterization can lead to improved patient management and outcomes.