Detecting referable glaucoma using artificial intelligence

Project Member(s): Golzan, M., Kennedy, P.

Funding or Partner Organisation: National Health & Medical Research Council (NHMRC - Ideas Grants)
National Health & Medical Research Council (NHMRC - Ideas Grants)

Start year: 2022

Summary: Glaucoma, a leading cause of irreversible blindness, is expected to affect 112 million people globally by 2040. In Australia, fifteen people are diagnosed with glaucoma every day, however at least as many cases remain undiagnosed. The high rate of undiagnosed cases is mainly due to the asymptomatic nature of the early stages of glaucoma. Accordingly, diagnosing glaucoma prior to the clinical onset of the disease can be a complex pathological challenge, which is highly dependent on trained specialists. It is therefore not surprising that an audit of undiagnosed patients has found that more than 59% of these cases had visited an eye-care provider within the previous 12 months. To address this challenge, we have developed an artificial intelligence (AI) model trained on retinal videos that has been able to achieve a sensitivity and specificity of well above 0.9 (max=1) for glaucoma detection, superior to any other method. Our novel and innovative AI model analyses time-sequence images of the retina to extract dynamic retinal vascular markers in addition to optic nerve head structural features for glaucoma detection. The success of our approach stems from our earlier research that has demonstrated that a range of dynamic physiological processes underpin glaucomatous pathology. In this project, we will establish the efficacy of our state-of-the-art model in detecting referable glaucoma in comparison to the gold standard (i.e., specialist grading). Once we have established that our model is comparable to the gold standard, findings from this project can be used as a clinical decision support tool in primary care settings for identifying at-risk individuals for referral to a specialist.

FOR Codes: Vision science, Artificial intelligence, Human Diagnostics

Acknowledgement of Country

UTS acknowledges the Gadigal people of the Eora Nation, the Boorooberongal people of the Dharug Nation, the Bidiagal people and the Gamaygal people, upon whose ancestral lands our university stands. We would also like to pay respect to the Elders both past and present, acknowledging them as the traditional custodians of knowledge for these lands.