AI-enabled mobile-based fundus imaging for diabetes screening - retinal imaging and artificial intelligence (AI)-driven oculomics are novel, promising, non-invasive tools for clinic and…

The Department of Health and Human Services National Institutes of Health Fogarty International Center awarded Emory University $129,531 on September 1, 2026, under the International Research and Research Training program (CFDA 93.989) to develop and validate artificial intelligence software for type 2 diabetes screening integrated into mobile-based fundus imaging.

The recipient will develop AI software capable of detecting type 2 diabetes using smartphone-based retinal images collected from approximately 120 people without and with diabetes, then integrate and beta-test the software in a sample of 60 patients. The work includes evaluation of the tool's human-centric usability for healthcare screeners operating the device in hospital settings. This Phase R21 award supports development and feasibility testing; a subsequent R33 phase is anticipated to field-test scalability and effectiveness. The AI module augments existing AI-integrated fundus imaging on mobile phones (Remidio Fundus-on-Phone, FDA-approved for diabetic retinopathy screening) developed through a research-commercial partnership between US and India-based entities. The research team is multidisciplinary, spanning the United States and India, with India selected as a validation site due to high type 2 diabetes prevalence and applicability of screening challenges to rural US populations and low-income countries.

Work is performed in Atlanta, Georgia. The period of performance runs through June 30, 2028. This is a project grant, an assistance type supporting research and research training to reduce global health disparities and foster international scientific partnerships.

AI-enabled mobile-based fundus imaging for diabetes screening - retinal imaging and artificial intelligence (AI)-driven oculomics are novel, promising, non-invasive tools for clinic and community-based screening for conditions such as type 2 diabetes (T2D), which often remains undiagnosed until major complications occur. Even when screening programs exist, individuals with newly diagnosed T2D often present with vascular damage including some degree of diabetic retinopathy (DR), showing the importance of screening to prevent disease progression. Identifying easy to use and disseminate, non-invasive solutions would benefit individuals in low resource settings or where healthcare penetration is low, such as rural US or low-income countries like india. India provides an ideal location to test T2D screening tools due to its high T2D prevalence, allowing rapid testing and development, while facing many of the screening challenges seen in the US. This group showed that AI-driven, mobile phone fundus imaging/oculomics has high sensitivity, specificity and accuracy for detecting T2D. Our multidisciplinary team from the US and india proposes to expand upon this early work and existing research-commercial partnership with a US/india-based company to develop, validate, and field test the addition of AI-driven T2D screening to existing AI-integrated fundus imaging on mobile phones developed for DR screening (remidio fundus-on-phone, US FDA approved for DR screening). In the R21 phase, we aim to: 1. Develop the AI software for T2D detection (using 120 existing and newly collected smartphone-based retinal images in people without and with T2D) and integrate and beta-test the software in a sample of 60 patients and 2. Evaluate, using mixed methods, the human-centric usability of the tool for T2D screeners (e.g., healthcare providers using the tool in a hospital setting). We will then test the feasibility, scalability, and effectiveness in a field-based study (R33 phase) by: 3. Comparing the diagnostic yield of T2D of AI-enabled smartphone-based fundus imaging compared to conventional, standard of care T2D screening in india (random capillary glucose testing followed by confirmatory fasting plasma glucose testing) as well as estimate the burden of DR among undiagnosed T2D cases; and 4. Evaluate, using mixed methods, the feasibility, scalability, and costing of AI-enabled smartphone-based fundus imaging compared with conventional T2D screening in community outreach centers. If successful, this innovative application of oculomics for T2D detection would bridge the screening GAP by providing a low-cost non-invasive screening tool that is easy to use, disseminate, and sustain, thereby improving early diagnosis of T2D and DR and reducing the burden of T2D and its complications.