You are here:  

• Print

PhD Studentship in Development of dynamic indexes of dairy cow welfare-resilience and improvement and extension of current health and welfare alerts from temperature and activity bolus

About the project

Accurate prediction of disease has the potential to transform livestock productivity, sustainability, and welfare. Behavioural and physiological changes in dairy cattle can signal key health and reproductive events such as oestrus, calving, subclinical ketosis, milk fever, mastitis, and acidosis. Among these, mastitis, lameness, and metabolic disorders such as ketosis are particularly significant and represent major welfare and economic challenges. Precision livestock technologies (PLTs) are increasingly being used to detect these conditions early and improve herd management (Besler et al., 2024).

Some health conditions, including clinical ketosis (Rasmussen et al., 2022), have relatively low prevalence, making accurate prediction challenging due to limited data. Recent advances in quantification algorithms have improved classification performance for such datasets (Carslake et al., 2021). To ensure reliable and robust health alerts, it is essential to account for both individual variability (e.g., body size) and environmental factors (e.g., farm differences). These challenges can be addressed using hybrid learning approaches that combine online and offline algorithms (Vázquez-Diosdado et al., 2019) and domain adaptation methods (Ahn et al., 2023).

Animal welfare is central to sustainable dairy farming. Cows experience multiple environmental and health stressors, such as disease and heat stress, which reduce fertility and productivity - pressures expected to intensify with climate change. Enhancing resilience, defined as the ability of animals to withstand and recover from disturbances (Colditz and Hine, 2016), is therefore crucial. While existing resilience indicators based on milk yield or activity data (Poppe et al., 2022) have shown limited predictive value, integrating a broader range of behavioural and physiological time-series data (e.g., activity, temperature, pH) offers significant potential.

Consumer demand for improved animal welfare is also increasing, yet current assessments often rely on manual, subjective methods such as the Qualitative Behaviour Assessment (QBA) (Dawkins, 2022). Insights from human studies indicate that long-term behavioural data can serve as reliable indicators of well-being (Voukelatou et al., 2020). PLTs, already used for health and productivity monitoring, can therefore be extended for welfare assessment (Molina et al., 2020). The SmaXtec bolus system, which continuously records physiological and behavioural parameters, provides an ideal platform for developing and validating such welfare indicators.

This PhD will develop short- and long-term welfare-resilience indices using activity and physiological data, while improving existing SmaXtec health alerts.

The project will involve:

• Algorithm improvement: Enhance and evaluate machine learning models for disease prediction.
• Index development: Create welfare-resilience indices informed by literature and farmer consultation, incorporating various metrics such as somatic cell count, antibiotic use, and behavioural and physiological measures. A short-term heat stress index will also be developed.
• Application: Use long-term indices to predict performance outcomes and monitor recovery and adaptability to stressors.

Research will be conducted at the Centre for Dairy Science and Innovation and collaborating dairy farms. The successful candidate will gain expertise in precision livestock technologies, data analytics, epidemiology, and social science, along with training in software/hardware integration and commercial implementation.

Research Environment

Research at the School of Veterinary Medicine and Science includes established world class research groups. 85% of our research is classed as "world-leading" (4*) or "internationally excellent" (3*) and our research collaborations and networks extend nationally and internationally.  Research undertaken at the School is relevant to both Veterinary Medicine and Science, One Health, and Comparative and Human Medicine.

Entry requirements

Eligibility:

Applicants should have a minimum of a 2.1 undergraduate degree or a minimum of a 2.2 degree and a Master’s degree. Subjects such as Computer Science, Data Science, Veterinary Science, Mathematics, Animal Science and similar subjects. 

This fully funded PhD, jointly supported by SmaXtec and the University of Nottingham School of Veterinary Medicine and Science, offers a unique opportunity to contribute to the development of next-generation animal health and welfare monitoring tools - bridging science, technology, and industry.

How to apply

Informal enquiries may be addressed to the principal supervisor: Jasmeet Kaler (Jasmeet.Kaler@nottingham.ac.uk)

Candidates should apply online and include a CV.  When completing the online application form, please select the School of Veterinary Medicine and Science, then PhD Veterinary Medicine and Science (36m) and, once submitted, send your student ID number to SV-PG-VET@exmail.nottingham.ac.uk

Any queries regarding the application process should be addressed to SV-PG-VET@nottingham.ac.uk

Interview date: 5th January 2026