Gait feature extraction in Parkinson's disease using low-cost accelerometers

Area of Science:

• Biomedical Engineering
• Neurology
• Movement Science

Background:

• Parkinson's disease (PD) is characterized by motor symptoms including gait disturbances like slowness, shuffling, and freezing of gait (FoG).
• Current methods lack comprehensive tools for fully characterizing these gait impairments in PD patients.
• Objective quantification of parkinsonian gait is crucial for effective management and research.

Purpose of the Study:

• To develop and evaluate a novel four-accelerometer system for detailed characterization of gait disturbances in Parkinson's disease.
• To establish a system capable of extracting a wider range of gait parameters beyond current clinical capabilities.
• To lay the groundwork for improved diagnostics and monitoring of PD-related gait issues.

Main Methods:

• Development of a hardware system utilizing four accelerometers.
• Implementation of an algorithm for automatic stride-by-stride signal epoching and analysis.
• Quantification of key gait parameters including velocity, stride time, stance/swing phases, support phases, and toe-off acceleration.
• Validation against visual inspection of video recordings.

Main Results:

• The developed algorithm successfully processed accelerometer data on a stride-by-stride basis.
• Key gait parameters were quantified, showing potential for detailed gait disturbance characterization.
• Preliminary results were presented from a Parkinson's disease patient and a healthy volunteer.

Conclusions:

• The four-accelerometer system shows promise for enhancing the characterization of parkinsonian gait disturbances, including freezing of gait (FoG) and gait asymmetries.
• Further improvements, such as incorporating time-frequency analysis for FoG detection, are planned.
• The system is slated for validation against a state-of-the-art 3D movement analysis system.