Leonor Corsino, MD, MHS, associate professor of Medicine and Population Health Sciences in the Division of Endocrinology, Metabolism, and Nutrition, has received a one-year pilot grant from the Duke Global Health Institute (DGHI). The study, initiated earlier this year, aims to evaluate an advanced artificial intelligence system, called DeepDKD Plus, designed to predict diabetic kidney disease (DKD) progression using retinal imaging in people with diabetes in the Dominican Republic.
Diabetes and its complications, including kidney disease and blindness, have increasingly become critical health challenges globally, particularly affecting low- and middle-income countries like the Dominican Republic. Recent epidemiologic data indicate diabetes prevalence in the adult Dominican population has reached an alarming 12.7%, ranking diabetes among the top four causes of death nationally. This underscores an urgent need for innovative, scalable solutions for early detection and management of diabetes-related complications, particularly in settings with limited health care resources.
The pilot study is designed around the hypothesis that retinal images analyzed by the AI-driven DeepDKD Plus system can effectively predict the onset and progression of diabetic kidney disease. This method has the potential to enable earlier intervention, reducing the health care burden while significantly improving patient outcomes.
Dr. Corsino, who has dedicated much of her research career to addressing disparities in diabetes care, emphasizes that "leveraging artificial intelligence is an essential strategy to mitigate health inequities, especially in low-resource settings." She adds, "Given the shortage of health care providers in these regions, technology like DeepDKD Plus could dramatically enhance clinical capabilities and patient care."
The pilot project builds upon longstanding international collaborations established by Dr. Corsino and her team. Since 2017, when she received her initial travel grant from DGHI, Dr. Corsino has developed strong partnerships with medical and research institutions in the Dominican Republic. In the last two years, she developed collaborations with investigators around the world working in AI and diabetes care.
For the current pilot, Dr. Corsino has assembled a robust, interdisciplinary team from Duke and collaborating international institutions. Key co-investigators include Bin Sheng, PhD, from Shanghai Jiao Tong University, who is leading the technical development of the DeepDKD Plus AI model, and Michelle Jiménez de Tavárez, MD, PhD, from Pontificia Universidad Católica Madre y Maestra (PUCMM) in Santiago, Dominican Republic, overseeing patient recruitment and data collection.
At Duke, collaborators include ophthalmologist Ramiro Maldonado, MD, assistant professor; nephrologist David Ortiz-Melo, MD, associate professor in the Department of Medicine; and John Bartlett, MD, professor of medicine and Global Health, serving in an advisory role. These partnerships highlight the project's uniquely multidisciplinary and multinational approach.
Over the next year, the team will enroll approximately 100 adult patients from two major academic hospitals in Santiago, DR, affiliated with PUCMM. Participants will undergo comprehensive baseline assessments, including clinical measurements, retinal imaging using portable fundus cameras, and laboratory evaluations to establish kidney function status. The collected data, managed securely through Duke’s REDCap platform, will then be analyzed by the DeepDKD Plus system to assess the tool’s predictive accuracy across three critical stages of kidney disease progression.
Dr. Corsino is optimistic about the potential outcomes of this study, noting its alignment with the DOM mission to advance equitable care and clinical innovation. “This pilot is a critical step toward validating tools that could transform diabetes management not only in the Dominican Republic but across other resource-limited settings globally,” she said.
Beyond its immediate goals, the study strategically positions the team for future grant opportunities, including potential larger-scale NIH-funded projects.
Reflecting on the significance of these collaborations, Dr. Corsino emphasized the broader implications of her research for global health equity. "At Duke, we are uniquely positioned to partner across disciplines and borders," she said. "These global connections allow us to test cutting-edge technology in ways that make tangible impacts on communities facing the highest burdens of diabetes and kidney disease."