Animals : an Open Access Journal from MDPI · 5 authors, 3 centres
This summary was generated by AI from a single paper. It has not been reviewed by a clinician and is not clinical advice. Verify against the source before acting on it.
This review examines precision livestock farming technologies for dairy calves, including automatic milk feeding systems, accelerometers, infrared thermography, and sound analysis. Most research focuses on detecting health problems and monitoring pain, but a consensus on behavioral indicators is lacking. Integrating multiple technologies with machine learning shows promise for improving calf welfare, though commercial applications are not yet available.
**Background:** Precision livestock farming (PLF) technologies are increasingly used to monitor and improve the health and welfare of dairy calves, which face high morbidity and mortality rates. This narrative review aims to analyze the state-of-the-art of technological applications in dairy calves, covering automatic milk feeding systems (AMFS), triaxial accelerometers, infrared thermography (IRT), image processing, heart rate monitors, ruminal boluses, location devices, sound analysis, multi-technological approaches, and machine learning. The review highlights the need for validation studies and commercial applications.
**Methods:** The authors conducted a narrative review of the scientific literature, focusing on studies published up to 2021. They searched for papers addressing PLF technologies in dairy calves, including those on health detection, pain monitoring, behavior assessment, and welfare evaluation. The review synthesizes findings from multiple studies, discussing the strengths, limitations, and research gaps for each technology.
**Key Results:** AMFS is the most extensively used technology, with studies showing that sick calves reduce milk intake, drinking speed, and unrewarded visits 2–4 days before diagnosis. For example, Conboy et al. reported a 63% reduction in daily milk intake in calves with bovine respiratory disease (BRD) and 57% in those with neonatal calf diarrhea (NCD). Triaxial accelerometers are used to monitor lying behavior, with studies showing that diarrheic calves had 64.8 min longer lying time one day before clinical identification. IRT has been used to detect BRD, NCD, and omphalitis; for instance, an increase in orbital temperature was observed 4–6 days before BRD symptoms, with sensitivity of 68.7% and specificity of 77.4%. However, IRT correlations with rectal temperature are weak (r = 0.16–0.50). Sound analysis for cough detection in BRD showed precision of 87.5% and specificity of 99.2% but sensitivity of only 50.3%. Multi-technological approaches, such as combining AMFS and accelerometers, improved disease prediction, with machine learning models achieving sensitivity of 0.54 and specificity of 0.95 for BRD prediction within a 3-day window.
**Clinical Implications:** The review underscores that PLF technologies can aid in early disease detection and pain assessment, potentially reducing morbidity and mortality in dairy calves. However, many technologies require further validation, and commercial applications are not yet proven effective. The integration of multiple sensors with machine learning algorithms offers the best potential for improving calf welfare, but human observation and decision-making remain essential. Future research should focus on developing automated systems for individual calf monitoring, especially during the early postnatal period, and on validating technologies for positive welfare indicators.