The development of early oesophageal cancer manifests with microscopic changes in the structure of the tissue that are invisible using current endoscopy techniques. We are developing a new endoscope that enables the acquisition of imaging data over a wide field of view at lower resolution (‘red flag’) combined with a computational zoom to provide a microscopic ‘optical biopsy’ image with high resolution in a narrow field of view. This enables the endoscopist to interrogate suspicious lesions more thoroughly without the need for complex bulky instrumentation or increased imaging examination time.
The technology in this endoscope uses optical fibres, which present a challenge as they scramble the phase and polarisation information present in the input coherent light . We are therefore unable to implement the technology in a clinical application without investment in deeper mathematical insights. This project aims to overcome the present challenge by casting the experimental setup as a highly non-trivial sparse inverse problem.
This project could appeal to a candidate with an interest in inverse problems, sparse regularization image processing and optimisation.
 G S D Gordon, J Joseph, S E Bohndiek and T D Wilkinson (2015) Single-pixel phase-corrected fiber bundle endomicroscopy with lensless focussing capability. J Lightwave Technol. 33 16 3419-3425