Technical note: Validation of sensor-recorded lying bouts in lactating dairy cows using a 2-sensor approach.

Area of Science:

• Animal Welfare Science
• Agricultural Technology
• Data Science in Agriculture

Background:

• Cow lying behavior (LB) is crucial for welfare assessment.
• Automated data loggers on hind legs can record LB.
• A duration threshold is needed to filter false records from leg movements.

Purpose of the Study:

• Validate LB records from IceQube sensors in lactating dairy cows.
• Determine an optimal threshold to differentiate true from false LB records.
• Implement a novel, time-efficient validation method.

Main Methods:

• Equipped 28 lactating cows with IceQube sensors on both hind legs for 6 days.
• Used a novel validation approach where sensors on each leg validated the other.
• Classified LB records based on assumptions of standing, simultaneous false records, and coinciding true records.

Main Results:

• False LB records constituted 4% of total records.
• A minimum LB duration threshold of 33 seconds achieved maximum accuracy (0.99), sensitivity (0.99), and specificity (0.98).
• Applying the 33s threshold minimally impacted daily lying time estimates but improved LB frequency and duration accuracy.

Conclusions:

• The 33-second threshold reliably distinguishes true from false LB records using IceQube sensors.
• This validation method is time-efficient and applicable across different sensors and conditions.
• Accurate LB detection is vital, particularly for identifying critical events like calving.