Foot gait time series estimation based on support vector machine.

Area of Science:

• Biomechanics
• Signal Processing
• Machine Learning

Background:

• Accurate stride interval time series estimation is crucial for gait analysis.
• Existing methods may lack precision in detecting key gait events like heel strikes.

Purpose of the Study:

• To propose a new algorithm for estimating stride interval time series from foot gait signals.
• To enhance the accuracy of heel strike detection using machine learning and morphological operations.

Main Methods:

• Utilizing a support vector machine (SVM) for detecting the beginning of heel strikes.
• Applying morphological operations to improve the accuracy of heel strike detection.
• Estimating stride interval time series by calculating backward differences of detected heel strikes.

Main Results:

• The proposed algorithm demonstrates accurate estimation of stride interval time series.
• Simulation results show estimation errors for the mean and standard deviation are on the order of 10(-4).

Conclusions:

• The developed algorithm provides a highly accurate method for stride interval time series estimation.
• This technique offers a reliable tool for quantitative gait analysis.