Researchers

Name: Zane Hartley

Current project titles: Cat Royale, Syngenta

Research description and interests: Zane Hartley is a Research Fellow in the School of Computer Science at the University of Nottingham, currently working with Somabotics on the Cat Royale project. His research background encompasses computer vision and generative AI, with a focus on agricultural applications, including 3D plant phenotyping and the development of synthetic data using diffusion models. Following his PhD, he has continued his research funded by Syngenta and was awarded the EPSRC Doctoral Prize in 2024.

Name: Nathan Mellor

Current project titles:  Delivering Sustainable Wheat - Hyperspectral image analysis for phenotyping and yield prediction

Research description and interests:  I’m currently working on a project in collaboration with field experimentalists at Rothamsted Research (Harpenden, UK) using UAV (drone) acquired hyperspectral image data of field grown wheat crops and a range of machine learning and deep learning techniques to extract and predict traits of interest such as yield and grain Nitrate content. In my previous project I developed content for the DataCampp project, a series of online courses designed to train bioscience professionals working in plant phenotyping basic skills and knowledge in image analysis, machine learning and deep learning. Previous research projects from my PhD and background in mathematical modelling have included producing 3D representations of crop root systems and modelling root-soil interactions, and vertex-based models of hormone distribution within plant roots.

Name: Janet Kok (Yong En)

Current project titles: Physics-informed Network for Optical Aberration Correction, Plant Root Segmentation

Research description and interests: My research in computer vision centers on developing robust, interpretable, and physically-grounded AI to solve complex challenges across healthcare, engineering, and the life sciences. By integrating underlying scientific principles into the models, my goal is to ensure they are reliable and can be trusted by domain experts to accelerate scientific discovery. As a member of the EPSRC InLightenUs research group, I have developed physics-informed neural networks that provide a generalisable and accurate correction for optical aberrations across diverse microscopy modalities and samples. My interdisciplinary experience extends to diverse applications, from building interpretable models for osteoarthritis diagnosis using Raman spectra and segmenting brain haemorrhages, to engineering a full-stack, AI-assisted annotation tool for plant phenomics. Furthermore, I leverage generative AI, specifically implementing diffusion models to design manufacturable cellular structures and static mixers for chemical engineering.

Name: Suhaib Shahid

Current project titles: External-to-Internal Correlation Modelling (E2ICM) for real-time MRI vocal tract synthesis

Research description and interests: My research focuses on developing advanced AI frameworks that bridge external visual data and internal physiological imaging to enable non-invasive modelling of anatomical structures. I am particularly interested in real-time MRI and deep learning methods that capture and correlate dynamic movements of the oral cavity during speech and mastication. My work explores generative and segmentation-based architectures, including GAN and diffusion models, alongside self-supervised approaches for temporal tracking and representation learning. These techniques aim to enhance the precision of internal anatomy reconstruction from external cues, contributing to the broader goal of multimodal integration in medical imaging. I am also investigating the translational applications of deep learning and MRI in dental care, speech research, and medical education, through collaborations with industry partners and the School of Medicine.