Validating a Precision Livestock Farming Method to Determine Individual, Daily Grazing Feed Intake
Authors:
Preprint
- 6 November 2026
- preprint
- Published by MDPI AG in Preprints.org
Abstract
Accurate measurement of individual dry matter intake (DMI) is fundamental to eval-uating feed efficiency and other performance traits in beef cattle but remains difficult un-der grazing conditions. The study objective was to externally validate a previously devel-oped long short-term memory (LSTM) model for predicting individual grazing DMI using independent cattle populations and grazing environments that differed from those used during model development. A total of 113 grazing beef heifers representing four cohorts were evaluated across three grazing performance tests conducted in West Virginia and Georgia during 2024. Individual DMI was predicted using the LSTM model from passively collected animal and environmental data (ML-DMI). Ground-truth DMI was inde-pendently estimated using titanium dioxide as an external marker and forage digestibility (TiO₂-DMI). Agreement between methods was evaluated using root mean square error (RMSE), matched-pairs analysis, Bland–Altman analysis, and a repeated-measures linear mixed model. Daily RMSE was 1.60 kg·d⁻¹ and decreased to 1.35 kg·d⁻¹ when DMI was av-eraged across validation days. Mean difference between methods was −0.097 kg·d⁻¹ (P = 0.45), and DMI-estimation method was not significant in the mixed model (P = 0.65). These results support the application of the LSTM model to independent cattle populations and dissimilar grazing environments.Keywords
All Related Versions
This publication has 40 references indexed in Scilit:
- Review: Genomic selection in the era of phenotyping based on digital imagesAnimal, 2025
- Using machine learning methods to predict dry matter intake from milk mid-infrared spectroscopy data on Swedish dairy cattleJournal of Dairy Research, 2023
- Diet, marker and fecal sampling method interactions with internal and external marker pairs when estimating dry matter intake in beef cattleLivestock Science, 2021
- Predicting feed intake using modelling based on feeding behaviour in finishing beef steersAnimal, 2021
- A review on the long short-term memory modelArtificial Intelligence Review, 2020
- Review: Synergy between mechanistic modelling and data-driven models for modern animal production systems in the era of big dataAnimal, 2020
- Evaluation of titanium dioxide and chromic oxide as digestibility markers in ponies fed alfalfa hay in relation to marker dosing frequencyAnimal, 2019
- Evaluation of external markers to estimate fecal excretion, intake, and digestibility in dairy cowsTropical Animal Health and Production, 2014
- Evaluation of titanium dioxide as an inert marker for estimating faecal excretion in grazing sheepAnimal Feed Science and Technology, 2009
- Excretion patterns of titanium dioxide and chromic oxide in duodenal digesta and feces of ewesSmall Ruminant Research, 2006