Background: Overweight, obesity, and type 2 diabetes (T2DM) increase the risk of vascular complications, including retinal vascular alterations. Artificial intelligence (AI)—based retinal screening enables early detection of microvascular damage, potentially contributing to appropriate cardiometabolic prevention strategies. We performed a realworld evidence observational study to assess cardiometabolic parameters as well as retinal vascular alterations using novel AI-based retinal imaging. Methods: Between September and December 2025, we recruited 301 consecutive adults with overweight,obesity, or T2DM; cardiometabolic assessment was performed, and all participants also underwent retinal screening with the Aireen AI system, distinguishing those who resulted positive (n = 42), negative (n = 204), or undetermined (n = 55) at the screening. Results: All patients who resulted positive at the screening were confirmed with a diagnosis of retinal alterations by the ophthalmologists. The presence of positive retinal screening (RS+) progressively increased with worsening fasting glycemia, glycated hemoglobin (HbA1c) levels, carotid intima–media thickness, and endothelial function (P < 0.001 for all). Multivariate logistic regression analysis revealed independent associations of RS+ with HbA1c [odds ratio (OR) 2.29, P < 0.001], carotid stenosis (OR 1.06, P < 0.001), fasting glucose (OR 1.02, P < 0.001), and the triglyceride/high-density lipoprotein cholesterol ratio (OR 2.19, P = 0.010). Conclusions: These findings suggest that AIbased retinal screening with advanced tools can effectively identify early microvascular damage in individuals with overweight, obesity, or T2DM, with RS+ prevalence reflecting glycemic burden and vascular dysfunction. Therefore, integrating AI retinal assessment into cardiometabolic prevention programs may enable timely interventions to prevent progression to vision-threatening complications and support comprehensive cardiovascular risk management.
Muratore, E., Maggio, V., Vaccaro, F., Patti, A.M., Giammanco, M., Di Falco, F., et al. (2026). Cardiometabolic Assessment in Patients with Overweight,Obesity, or Type 2 Diabetes Undergoing Artificial Intelligence–Based Retinal Screening: A Real-World Evidence Study. METABOLIC SYNDROME AND RELATED DISORDERS [10.1177/15578518261496].
Cardiometabolic Assessment in Patients with Overweight,Obesity, or Type 2 Diabetes Undergoing Artificial Intelligence–Based Retinal Screening: A Real-World Evidence Study
Elena Muratore;Viviana Maggio;Francesca Vaccaro;Angelo Maria Patti;Manfredi Giammanco;Felicia Di Falco;Mario Manunta;Vincenzo Geraci;Marcello Ciaccio;Luisa Agnello;Marco Giammanco;Vincenza Elena Bonfiglio;Manfredi Rizzo
2026-10-12
Abstract
Background: Overweight, obesity, and type 2 diabetes (T2DM) increase the risk of vascular complications, including retinal vascular alterations. Artificial intelligence (AI)—based retinal screening enables early detection of microvascular damage, potentially contributing to appropriate cardiometabolic prevention strategies. We performed a realworld evidence observational study to assess cardiometabolic parameters as well as retinal vascular alterations using novel AI-based retinal imaging. Methods: Between September and December 2025, we recruited 301 consecutive adults with overweight,obesity, or T2DM; cardiometabolic assessment was performed, and all participants also underwent retinal screening with the Aireen AI system, distinguishing those who resulted positive (n = 42), negative (n = 204), or undetermined (n = 55) at the screening. Results: All patients who resulted positive at the screening were confirmed with a diagnosis of retinal alterations by the ophthalmologists. The presence of positive retinal screening (RS+) progressively increased with worsening fasting glycemia, glycated hemoglobin (HbA1c) levels, carotid intima–media thickness, and endothelial function (P < 0.001 for all). Multivariate logistic regression analysis revealed independent associations of RS+ with HbA1c [odds ratio (OR) 2.29, P < 0.001], carotid stenosis (OR 1.06, P < 0.001), fasting glucose (OR 1.02, P < 0.001), and the triglyceride/high-density lipoprotein cholesterol ratio (OR 2.19, P = 0.010). Conclusions: These findings suggest that AIbased retinal screening with advanced tools can effectively identify early microvascular damage in individuals with overweight, obesity, or T2DM, with RS+ prevalence reflecting glycemic burden and vascular dysfunction. Therefore, integrating AI retinal assessment into cardiometabolic prevention programs may enable timely interventions to prevent progression to vision-threatening complications and support comprehensive cardiovascular risk management.| File | Dimensione | Formato | |
|---|---|---|---|
|
Cardiometabolic Assessment in Patients with Overweight.pdf
Solo gestori archvio
Tipologia:
Versione Editoriale
Dimensione
535.88 kB
Formato
Adobe PDF
|
535.88 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.


