Endomicroscopy is an emerging medical imaging modality that facilitates the acquisition of in vivo and in-situ optical biopsies, assisting fast diagnostic and potentially therapeutic interventions. To date, real-time endomicroscopy has been dominated and limited to intensity mode imaging due to existing detector technology. This limitation has now been overcome by a Fluorescent Lifetime Imaging (FLIM) system called Kronoscan, by incorporating both intensity and lifetime imaging. This breakthrough technology will enable multidimensional high-content real-time sensing and imaging of dynamic biological processes. These systems are now poised for disruptive healthcare impact, and a key ambition of the research will be to pave the way for subsequent clinical and commercial impact.

This project will use image processing and machine learning techniques for further developing the FLIM platforms with two key aims: to improve (1) image reconstruction and quantification of samples for assessment; and (2) monitoring of drug-target engagement. The student may build on various techniques for image reconstruction and bacteria detection, including computationally efficient Bayesian estimation and deep learning methods.

The project will be undertaken in partnership with GlaxoSmithKline (GSK), who have the eventual aim of applying high-resolution ultra-sensitive microscopic imaging to the evaluation of drug action, and the National Physical Laboratory (NPL), who will provide expertise and guidance on metrology and uncertainty considerations and practices applied to data processing, analysis and machine learning. The student will also be enrolled in the Postgraduate Institute for Measurement Science and receive training and co-supervision from NPL scientists.

For details of this 4-year PhD project with Dr James Hopgood, Dr Neil Finlayson, and Prof Kev Dhaliwal at the University of Edinburgh, go to https://www.eng.ed.ac.uk/studying/postgraduate/research/phd/machine-learning-techniques-evaluating-disease-and-drug

The deadline for application is 31 December 2020