An analysis of greyhound gait using wavelets

Area of Science:

• Biomechanics
• Signal Processing
• Animal Locomotion

Background:

• Gait analysis in quadrupedal animals is crucial for understanding locomotion and diagnosing neurological or orthopedic conditions.
• Traditional methods may not fully capture subtle changes in joint kinematics during locomotion.
• Wavelet transform offers a powerful tool for analyzing non-stationary time-series data, such as animal movement.

Purpose of the Study:

• To develop and validate a novel method for analyzing quadrupedal animal gait using discrete wavelet transform (DWT).
• To apply this method to detect and quantify gait abnormalities in greyhounds with induced tibial nerve paralysis.
• To compare the energy distribution of wavelet components between normal and abnormal gait patterns.

Main Methods:

• Collected time-series kinematic data of hindlimb joints from three greyhounds under normal conditions and with induced tibial nerve paralysis.
• Applied discrete wavelet transform (DWT) to decompose joint trajectory data into different frequency components.
• Compared the energy contribution of corresponding wavelet components between normal and abnormal gait datasets.

Main Results:

• The wavelet transform successfully decomposed complex gait data into distinct frequency components.
• Significant differences in the energy distribution across wavelet levels were observed between normal and paralyzed hindlimb gaits.
• The method effectively detected and quantified gait abnormalities, correlating with the severity of induced paralysis.

Conclusions:

• Wavelet analysis provides a sensitive and quantitative method for assessing quadrupedal animal gait.
• This technique can identify and characterize kinematic alterations associated with gait pathologies.
• The approach holds potential for early diagnosis and monitoring of locomotion disorders in animals.