Recent publications by SoVS staff include:
1 – What makes good healthcare AI?: Lessons from diabetic retinopathy screening
Eyeonoptics magazine
David Squirrell, Ehsan Vaghefi, Aan Chu
Artificial Intelligence is reforming many fields of medicine, but optometry and ophthalmology are particularly affected by new wave of AI technologies. While the AI research in optometry and ophthalmology has been progressing very rapidly, very few have been successful to become clinically applicable. There are several technical, ethical and clinical barriers for transition of research AIs into clinical tool. In this article we are investigating some of the more important barriers of clinical application of AI in optometry and ophthalmology
Click here to read the paper.
2 -Multi-Modal Retinal image analysis via deep learning for the diagnosis of intermediate dry age related macular degeneration: A feasibility study
Journal of Ophthalmology
Ehsan Vaghefi, Sophie Hill, Hannah Kersten, David Squirrell
AI is most often (%99 of the time) is trained and then used on a single mode of data e.g. one imaging modality or signal. However, the actual clinical workflow of disease diagnosis is very different. i.e. the clinician considers input from several imaging modalities and also takes into account patient’s metadata, before making a clinical decision. In this paper we tested a proof of concept AI architecture, where we trained and validated our AI on three “combined” modalities (OCT\ OCTA and fundus camera). Although a prototype, we showed that a “multi-context” AI is much more accurate than a single mode AI, in diagnosis of an eye disease (i.e. macular degeneration)
Click here to read the paper.
3 -Effective refractive error coverage: An eye health indicator to measure progress towards universal health coverage.
Ophthalmic and Physiological Optics|
Ian McCormick, Islay MacTaggart, Andrew Bastawrous, Matthew J Burton, Jacqueline Ramke.
This paper defines and discusses the measurement of the effective refractive error coverage indicator, which is one of the two indicators prioritised for national and global monitoring in the World Health Organization’s first World Report on Vision published in late 2019.
Click here to read the paper.