Artificial intelligence in ophthalmology: from data to algorithm and real-world application

Summary This project aims to further evolve artificial intelligence technology to develop and validate a clinical decision system that can predict disease outcomes and prognosis, as well as help clinicians decide on treatment options, based on real-world multi-modality clinical data.

Project Details

Funded by NHMRC Investigator Grant from 2020-2025, this project aims to further evolve artificial intelligence technology to develop and validate a clinical decision system that can predict disease outcomes and prognosis, as well as help clinicians decide on treatment options, based on real-world multi-modality clinical data. In this project, Professor Mingguang He will conduct a data linkage project to establish a cohort of people with ocular imaging data by retrieving historic clinical records from the eye care service providers and then link these people with Medicare claim and hospitalisation data. This unique cohort will help develop algorithm using ocular imaging as biomarkers for systemic disease prediction.Using the latest AI technology, a Smart Clinical Query System will be developed as a tool to answer open questions that clinicians may ask in clinical practice. Initially, this system will take one disease domain, for example glaucoma, and will then be further expand to other diseases using the same framework and strategies when validation is proven. The system will be validated in ophthalmology clinical practice settings to assess clinical accuracy, reproducibility against eye specialists, as well as the real-world impact and cost-effectiveness by comparing with conventional clinical management decisions.