Real-time pixel-level semantic interpretation of Intraoperative Optical Coherence images for guided regenerative therapy delivery

Lead Supervisor
Dr Christos Bergeles
Reader
School of Biomedical Engineering & Imaging Sciences
King’s College London
christos.bergeles@kcl.ac.uk

Co-supervisor
Prof Tom Vercauteren, Dr Lyndon Da Cruz

Project Details

Gene and cellular therapies are emerging transformative treatments for blinding retinal diseases. Treatment delivery into delicate retinal layers, some as thin as 10-20um, will be guided by Intraoperative Optical Coherence Tomography (iOCT) microscopes, a modality microscopically imaging not only the retinal fundus, but also the “hidden” subretinal layers where therapeutics should be delivered. This PhD project will computationally augment iOCT imaging through incorporation of real-time artificial intelligence (AI) in the acquisition and processing pathways towards the creation of a robust navigation system that guides therapy implantation.

The student will develop deep learning models for real-time pixel-level semantic understanding of iOCT images. While extensive work has been carried out on segmentation and interpretation of pre-operatively acquired OCT volumes, including retinal layer delineation and pathology identification, there is almost no work on the analysis of iOCT images. iOCT images have relatively low SNR, less spatial resolution, and must be interpreted in real time. The student will be at a unique position to develop compact powerful artificial neural networks to achieve this overarching objective given the team’s expertise on pixel level semantic interpretation of video streams, as well as the extensive datasets that have been collected from a variety of vitreoretinal surgical interventions.

Keywords

Surgical vision, surgery and intervention