Integrating Physiological Biomarkers into Precision Livestock Farming for Laying Hens: Methodological Challenges and Future Research Directions

Abstract

Continuous environmental, production, locational, image, and acoustic data have increasingly been collected from laying-hen farms through precision livestock farming (PLF) systems. However, sensor-derived classifications have often been validated against human-assigned behavioral or appearance labels rather than independently measured physiological states. In this narrative review, the methodological requirements for using physiological indicators as reference labels in PLF models for laying hens were examined, considering blood biochemistry, corticosterone concentration, the heterophil-to-lymphocyte ratio, body temperature, and immune or disease indicators alongside structured farm-level data and unstructured video, thermal, and acoustic data. Direct sensor-to-reference comparison studies numbered only seven and predominantly used rectal or cloacal temperature, peripheral temperature, clinical trajectories, or disease outcomes. No study identified in this review developed and externally validated a composite label combining blood biochemistry, the heterophil-to-lymphocyte ratio, corticosterone, and immune or pathogen indicators in laying hens. Major barriers were considered to include sensor–biomarker time-scale mismatches, sampling-induced label noise, and inconsistent units of analysis across individual, flock, and farm levels. It is suggested that future studies incorporate individual identification, biomarker-specific temporal alignment, multimodal composite labeling, and independent farm-level validation, so that PLF for laying hens can progress from sensor-centered detection toward biologically validated decision support linked to management actions.

The authors' abstract, as published at the source. Animals, 2026 · DOI ↗