Evaluation of Artificial Intelligence Algorithms for the Classification of Diabetic Retinopathy and Diagnosis of Hypertension in Multiethnic Brazilian Patients Using Portable Retinal Camera Images

• Grant number: 25/03562-7
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: June 01, 2025
• End date: May 31, 2026
• Field of knowledge: Interdisciplinary Subjects
• Principal Investigator: Caio Vinicius Saito Regatieri
• Grantee: Andre Kenzo Aragaki
• Host Institution: Escola Paulista de Medicina (EPM). Universidade Federal de São Paulo (UNIFESP). Campus São Paulo. São Paulo , SP, Brazil

Abstract

This project evaluates the accuracy and concordance of artificial intelligence (AI) algorithms in classifying diabetic retinopathy (DR) and diagnosing arterial hypertension (AH) in Brazilian patients from diverse ethnic backgrounds using portable retinal camera images. The study aims to validate AI models, identify ethnic and demographic biases, and propose solutions to promote greater equity in the diagnosis and treatment of these conditions. Retinal images will be sourced from a database and analyzed by machine learning and deep learning algorithms. The models' performance will be compared to human specialists using metrics such as sensitivity, specificity, and area under the ROC curve. This work seeks to generate scientific evidence to support the adoption of accessible technologies for public health screening, particularly in underserved populations, contributing to reducing inequalities and improving ophthalmological care in Brazil. (AU)