Machine Learning for the Prediction of Subclinical Lesions in Small Cerebral Vessels

• Grant number: 25/01366-6
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: April 01, 2025
• End date: March 31, 2026
• Field of knowledge: Health Sciences - Collective Health - Epidemiology
• Principal Investigator: Alexandre Dias Porto Chiavegatto Filho
• Grantee: Marianna Gerardo Hidalgo Santos Jorge Leite
• Host Institution: Faculdade de Saúde Pública (FSP). Universidade de São Paulo (USP). São Paulo , SP, Brazil

Abstract

Subclinical lesions in small vessels of the brain are pathological alterations that, although they do not cause obvious clinical symptoms, can be detected by advanced neuroimaging techniques. These lesions include cerebral microinfarcts, lacunes, white matter hyperintensities, microhemorrhages, and enlarged perivascular spaces, all of which are associated with an increased risk of cognitive decline, stroke, and mortality. Early identification of these lesions and appropriate management of risk factors, such as hypertension, are crucial to mitigate the progression of these lesions and their clinical consequences. Previous studies have explored the use of antiplatelet agents for the management of these lesions, but there is a significant gap in the understanding of the predictors of these lesions over time. This project proposes the use of machine learning algorithms to predict the progression of subclinical lesions. The research will use data from four waves of the Longitudinal Study of Adult Health (ELSA-Brasil), a multicenter cohort study involving 15,000 employees from six public higher education and research institutions in the Northeast, South, and Southeast regions of Brazil. Clinical variables and lifestyle scores will be included, and machine learning algorithms such as random forest and XGBoost will be applied to assess their performance in predicting the progression of subclinical lesions.