Gait Monitoring and Analysis: A Mathematical Approach

Area of Science:

• Gerontology
• Neurology
• Biomedical Engineering

Background:

• Gait abnormalities are prevalent in the elderly and Parkinson's disease (PD) patients, increasing fall risk and reducing mobility.
• Effective monitoring of gait is vital for understanding disease progression, early detection of motor impairments, and tailoring rehabilitation.
• Early identification of gait irregularities enables timely interventions, potentially delaying severe motor symptoms and improving patient outcomes.

Purpose of the Study:

• To investigate gait patterns in older adults with chronic diseases and Parkinson's disease.
• To develop and validate a novel metric for assessing "gait quality" using wearable sensor data.

Main Methods:

• Utilized wearable devices to monitor gait in approximately 50 participants (20 with PD, 30 with chronic diseases) over at least six months.
• Collected accelerometer data at a rate of 25 samples per second per device.
• Proposed a novel "gait quality" metric derived from entropy analysis following Fourier transformation of sensor data.

Main Results:

• Successfully collected extensive gait data from a diverse cohort of older adults.
• Developed a quantifiable metric for "gait quality" based on signal entropy and Fourier analysis.
• The proposed metric offers a novel approach to objectively assess gait characteristics.

Conclusions:

• The developed gait quality metric shows promise for objective assessment in clinical settings.
• This approach can support early detection of motor deficits and personalized treatment strategies for aging populations and those with Parkinson's disease.
• Wearable technology combined with advanced signal processing offers a powerful tool for gait analysis and patient management.