Table 1 Summary of studies in the current literature (neurodegenerative diseases)
From: Prognostic potentials of AI in ophthalmology: systemic disease forecasting via retinal imaging
Author | Study year | Key findings | Retinal biomarker | Dataset | Total No. of participants | Adjusted variables | Study design |
|---|---|---|---|---|---|---|---|
Cheung et al. [12] | Between August 2010 and December 2018 | Narrower retinal arteriolar calibre and wider retinal venular calibre were associated with an increased risk of cognitive decline and incident dementia. | SIVA-DLS | Participants from National University Hospital and St Luke’s Hospital in Singapore | 491 | Age Gender Ethnicity Education Cerebrovascular disease status Hypertension Hyperlipidaemia Diabetes Smoking | Prospective cohort study |
Hu et al. [13] | Between 2006 and 2010 | Each one-year increase in retinal age gap was independently associated with a 10% increase in the risk of incident PD. | Retinal age gap | UK Biobank | 35,834 | Age Gender Ethnicity Townsend deprivation Smoking status Drinking status Obesity Physical activity History of stroke History of diabetes mellitus Hypertension Use of psychotropic Medication | Prospective cohort study |
Wagner et al. [14] | Between 1st January 2008 and 1st April 2018 | Patients with AMD and DR were at a higher risk of developing AD compared to those without these eye diseases. These findings suggest that ophthalmic imaging (e.g., OCT and fundus photography) could be used as non-invasive tools for early detection and monitoring of AD. | Retinal photographs | Moorfields Eye Hospital in London, UK | 353,157 | Age Gender Ethnicity | Retrospective cohort study |
Cheung et al. [15] | - | The DL model achieved high accuracy in detecting AD in testing datasets, even when the datasets were from different populations and had different imaging protocols. The system could be used as a non-invasive, low-cost, and scalable tool for AD screening and monitoring. | Retinal photographs | MACC Novel Retinal Imaging Biomarkers for Cognitive Decline, Hong Kong Queen’s University of Belfast, UK SEED HKCES University of Hong Kong Volunteer Cohort | 3,888 (648 with AD, 3240 without AD) | Age Gender Presence of hypertension Presence of diabetes | Retrospective multicenter case control study |