Smoothing centre-of-mass tracking to quantify body oscillations and ambulation in broilers

Area of Science:

• Animal Science
• Computational Biology
• Animal Welfare

Background:

• Lameness is a significant welfare concern in broiler chickens.
• Current lameness assessment methods can be invasive and stressful for birds.
• Automated, non-invasive techniques are needed for objective broiler welfare evaluation.

Purpose of the Study:

• To develop a computational approach for quantifying broiler ambulation.
• To explore correlations between smoothed movement metrics and body weight.
• To establish a non-invasive method for assessing broiler mobility and welfare.

Main Methods:

• Broilers were video-recorded in an open field test at 28 days of age.
• Centre-of-mass tracking and zigzag trajectory analysis were performed using specialized software.
• Movement data were smoothed using a moving average, and step detection algorithms were applied.

Main Results:

• Smoothing broiler trajectories reduced estimated distance and speed significantly.
• Computed step counts were lower than observer counts, indicating potential underprediction.
• Strong positive associations were found between raw and smoothed movement metrics and observer vs. computed steps.

Conclusions:

• The computational approach shows potential for quantifying broiler ambulation non-invasively.
• Smoothing movement data may mitigate the impact of body oscillations on metrics.
• Further refinement of the step-detection method is necessary to improve accuracy for broiler welfare assessment.