Multimodal digital sensing of early-life laying hens: a pilot study integrating thermal, acoustic, optical-flow and environmental data.

Abstract

Early-life development profoundly shapes long-term welfare in laying hens, yet monitoring remains constrained by subjective assessment and fragmented single-modality tools. This pilot study evaluated the technical feasibility of a stratified multimodal sensing approach: thermal imaging and environmental monitoring across all five rooms (= 150 Lohmann LSL-Lite chicks) vs. detailed audio and video analyses limited to one representative room (= 30 birds) to manage annotation workload, from hatch to 20 weeks. One hundred fifty Lohmann LSL-Lite chicks were housed across five controlled rooms; thermal and environmental data were collected system-wide, whilst detailed audio and video analyses focused on one representative room to manage annotation workload. Weekly aggregated features included head and foot surface temperatures, acoustic spectral descriptors, optical-flow movement metrics around caretaker entry, and ambient conditions. Thermal imaging revealed age-related increases and stabilization of peripheral temperatures, with foot temperature showing a pronounced developmental effect (&#x3b7;= 0.51). Acoustic features shifted systematically across weeks (all< 0.001), consistent with vocal maturation. Optical-flow analysis revealed strong early reactivity to caretaker presence that declined markedly with development (early weeks 5-10 vs. late weeks 11-20:= 28.12,= 0.00126).-score-normalized multimodal trajectories and Pearson correlation analysis (Benjamini-Hochberg FDR,< 0.05) demonstrated strong within-modality consistency (= 0.85-0.96) and selective associations between environmental humidity and acoustic features (= 0.65-0.70), whilst thermal, acoustic, and behavioral domains remained largely independent. This descriptive pilot-thermal and environmental data from all rooms, behavior and vocalization from one cohort-establishes baseline multimodal developmental patterns and validates parallel sensing as a foundation for future welfare-relevant monitoring in precision poultry farming.