The objectives of this study were to develop instrumentation for measuring feed intake and forestomach motility for individually housed cattle and create data analysis algorithms to characterize feeding behavior and reticuloruminal contractions. Feed bunks were mounted onto S-beam load cells and suspended outside the animal stalls. Load cells were connected to a data logger which recorded bunk weight at 1-min intervals. Thus, periods of feed disappearance from the bunk could be equated to meals. A meal detection script was developed in MATLAB to import the weight data, filter the weight to reduce noise, detect meals by evaluating the difference between consecutive measurements, and export the meal start time, duration, and size of each meal. Evaluation of the meal detection script by visual review resulted in a combined error rate of 1.8%. A water-filled (2 L) balloon attached to a catheter was placed into the ventral sac of the rumen through the rumen cannula and connected to a disposable pressure transducer. The transducers were connected to a PowerLab data acquisition system via bridge amplifiers, and the ruminal pressure signal was visualized in real-time using LabChart computer software. A script was prepared in MATLAB for filtering pressure data, detecting contractions using the findpeaks function within the Signal Processing Toolbox of MATLAB, and evaluating contraction amplitude, duration, and frequency. Evaluation of the ruminal contraction script by visual review resulted in a combined error rate of 4%. After successful evaluation of the algorithms, an experimental application of these systems characterized meals and ruminal contractions for 8 animals in response to an increase in grain in the diet. This trial produced average daily results for feeding behavior and ruminal motility that were within the range of other published studies. With the low error rate and biologically acceptable values, the instrumentation and data analysis algorithms are appropriate means of characterizing feeding behavior and ruminal motility. These systems and algorithms could have important applications for ruminant physiology and behavior research.
|Journal||Computers and Electronics in Agriculture|
|State||Published - Dec 2019|
Bibliographical noteFunding Information:
The authors would like to thank Kirk Vanzant and Lauren Clark of the University of Kentucky C. Oran Little Research Center Beef Unit, as well as Kyle McLean (Ruminant Nutrition Post-Doctoral Scholar) and Winston Lin (Ruminant Nutrition Lab Supervisor) for assistance mounting the equipment and conducting trials of the recording systems.
© 2019 Elsevier B.V.
Copyright 2019 Elsevier B.V., All rights reserved.
- Feed disappearance
- Feeding behavior
- Rumen motility
ASJC Scopus subject areas
- Agronomy and Crop Science
- Computer Science Applications