Quantitative measurement of Parkinsonian gait from walking in monocular image sequences using a centroid tracking...

Area of Science:

• Biomedical Engineering
• Neuroscience
• Movement Science

Background:

• Parkinson's disease (PD) involves the degeneration of dopaminergic neurons, leading to progressive gait abnormalities.
• Accurate gait assessment is crucial for monitoring PD progression and treatment efficacy.
• Existing gait analysis methods may be limited by cost and space constraints in clinical settings.

Purpose of the Study:

• To develop and validate a portable, cost-effective method for assessing Parkinsonian gait using monocular video sequences.
• To quantify kinematic gait parameters in individuals with PD and compare them to healthy controls.
• To enhance gait analysis for PD in resource-limited environments like small clinics or homes.

Main Methods:

• A portable gait analysis system was developed using monocular image sequences and a centroid tracking algorithm.
• Two joint markers were attached to the leg (fibula head and lateral malleolus) of participants.
• Twelve healthy subjects and twelve mild PD patients were recorded walking naturally using a single digital video camera.

Main Results:

• Stride length and walking velocity were significantly reduced in PD patients not taking medication compared to those on medication and controls.
• Step length and swing velocity during gait initiation were also lower in PD patients without medication.
• The proposed method demonstrated a high correlation with the GAITRite® system, confirming its reliability.

Conclusions:

• The proposed portable video-based method provides an inexpensive, accessible, and reliable means for quantitative gait analysis in Parkinson's disease.
• This technique has the potential for widespread clinical and research application, including in home settings.
• The method shows promise for assessing gait in other neuromuscular conditions like stroke and traumatic brain injury.