Dr Giuseppe Di Caprio, University of Strathclyde; Dr Simon Cameron, Queen's University Belfast; Gaetano Di Caterina, University of Strathclyde; Andrew Smith, ABC Dairy Ltd / ASTB Ltd; Treenie Bowser, The Dairy Vet Ltd / ASTB Ltd.

This PhD project places artificial intelligence and computational modelling at the heart of next-generation mastitis diagnostics for sustainable dairy farming. Rather than focusing primarily on hardware development, the research targets the core scientific challenge: how to extract clinically meaningful information from complex optical data using advanced machine learning.
Mastitis remains the most common and costly disease in dairy cattle, driving significant antimicrobial use (AMU). Rapid, accurate on-farm diagnostics are essential to enable targeted treatment and reduce unnecessary antibiotics. While imaging flow cytometry and microfluidic platforms are now mature and can be implemented using modular, plug-and-play engineering solutions, the real bottleneck lies in robust, generalisable algorithms capable of interpreting high-dimensional image data under real-world farm conditions.
The student will develop advanced deep learning models to classify milk samples from healthy and infected cows, differentiating pathogen types directly from image streams. A central innovation will be the exploitation of diffraction-informed features to detect and classify bacteria that are near or below optical resolution limits.
A comprehensive, well-annotated digital library of milk images will be constructed, supported by access to isolated bacterial strains through collaboration with the ASTB Ltd. The developed algorithms will be benchmarked against existing biochemical and PCR-based diagnostic methods used on farms.
The student will gain strong expertise in machine learning, computational imaging, uncertainty quantification, and applied AI in agriculture, contributing to reduced antimicrobial overuse and advancing sustainable, data-driven dairy systems.