Daniel Essien

Yashan Dhaliwal, Daniel Essien, Suresh Neethirajan

Early-life development strongly influences long-term welfare in laying hens, yet monitoring remains limited by subjective assessment and single-modality tools. This pilot study evaluated the feasibility of a multimodal sensing framework integrating thermal imaging, acoustic recording, optical-flow-based video analysis, and environmental monitoring to characterize physiological and behavioural development 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, while detailed audio and video analyses focused on one representative room. Weekly aggregated features included head and foot surface temperatures, acoustic spectral descriptors, optical-flow movement responses to caretaker entry, and ambient conditions. Thermal imaging showed age-related increases and stabilization of peripheral temperatures, with foot temperature exhibiting a strong developmental effect (eta squared = 0.51). Acoustic features changed systematically across weeks (p < 0.001), consistent with vocal maturation. Optical-flow analysis revealed pronounced early reactivity to caretaker presence that declined with age (weeks 5 to 10 versus 11 to 20: t = 28.12, p = 0.00126). Z-score-normalized multimodal trajectories and correlation analysis (false discovery rate q < 0.05) showed strong within-modality consistency (r = 0.85 to 0.96) and selective associations between humidity and acoustic features (r = 0.65 to 0.70), while thermal, acoustic, and behavioural domains remained largely independent. This pilot establishes baseline multimodal developmental patterns and supports parallel sensing for welfare-relevant monitoring in precision poultry farming.

Daniel Essien, Suresh Neethirajan

The future of poultry production depends on a paradigm shift replacing subjective, labor-intensive welfare checks with data-driven, intelligent monitoring ecosystems. Traditional welfare assessments-limited by human observation and single-sensor data-cannot fully capture the complex, multidimensional nature of laying hen welfare in modern farms. Multimodal Artificial Intelligence (AI) offers a breakthrough, integrating visual, acoustic, environmental, and physiological data streams to reveal deeper insights into avian welfare dynamics. This investigation highlights multimodal As transformative potential, showing that intermediate (feature-level) fusion strategies achieve the best balance between robustness and performance under real-world poultry conditions, and offer greater scalability than early or late fusion approaches. Key adoption barriers include sensor fragility in harsh farm environments, high deployment costs, inconsistent behavioral definitions, and limited cross-farm generalizability. To address these, we introduce two novel evaluation tools - the Domain Transfer Score (DTS) to measure model adaptability across diverse farm settings, and the Data Reliability Index (DRI) to assess sensor data quality under operational constraints. We also propose a modular, context-aware deployment framework designed for laying hen environments, enabling scalable and practical integration of multimodal sensing. This work lays the foundation for a transition from reactive, unimodal monitoring to proactive, precision-driven welfare systems that unite productivity with ethical, science based animal care.