Factores clínicos asociados a la retinopatía del prematuro en neonatos tamizados con inteligencia artificial mediante VART Y RETCAM en una clínica de Barranquilla (2023–2024)
No hay miniatura disponible
Fecha
2025
Autores
Estupiñán Bucheli, María Isabel
Quintero Arciniegas, Laura
Título de la revista
ISSN de la revista
Título del volumen
Editor
Ediciones Universidad Simón Bolívar
Facultad de Ciencias de la Salud
Facultad de Ciencias de la Salud
Resumen
Introducción: La retinopatía del prematuro ( ROP) es una de las principales causas
de ceguera infantil prevenible, especialmente en países en desarrollo. En
Barranquilla, donde la atención neonatal ha mejorado, persiste la necesidad de
actualizar los datos sobre su incidencia, como también sobre las herramientas
diagnósticas y terapéuticas más utilizadas. En este contexto, la inteligencia artificial
(IA) se ha propuesto como una alternativa para mejorar el tamizaje oportuno y
reducir la carga sobre el especialista. Objetivo: Determinar los factores clínicos
asociados a la presencia y severidad de la retinopatía del prematuro en neonatos
tamizados con inteligencia artificial mediante VART y RetCam en una clínica de
Barranquilla durante el periodo 2023–2024. Materiales y métodos: Estudio
observacional, transversal retrospectivo, con enfoque analítico, realizado entre
enero de 2023 y diciembre de 2024. Se incluyeron 596 ojos de recién nacidos
prematuros tamizados mediante el sistema RetCam y evaluados con inteligencia
artificial (VART) en siete UCIN de Barranquilla. Se aplicaron análisis estadísticos
descriptivos, pruebas inferenciales y regresión logística para identificar los factores
asociados a la ROP. Resultados: La prevalencia de ROP fue del 19.5%. La edad
gestacional promedio fue de 28.6 semanas y el peso al nacer de 1263.7 g. Se
encontró una asociación significativa entre menor edad gestacional y bajo peso con
la aparición de ROP (p<0.001). El 84.4% de los casos con ROP recibió tratamiento,
siendo el láser la modalidad más utilizada (64.2 %), seguido de combinaciones con
Anti-VEGF (26.5%). La IA empleada en el sistema VART demostró una sensibilidad
del 80% para el tamizaje de ROP.
Introduction: Retinopathy of prematurity (ROP) is one of the leading causes of preventable childhood blindness, particularly in developing countries. In Barranquilla, where neonatal care has improved, there remains a need to update data on its incidence, as well as on the most commonly used diagnostic and therapeutic tools. In this context, artificial intelligence (AI) has been proposed as an alternative to improve timely screening and reduce the burden on specialists. Objective: To determine the clinical factors associated with the presence and severity of retinopathy of prematurity in neonates screened with artificial intelligence using VART and RetCam at a clinic in Barranquilla during the 2023–2024 period. Materials and Methods: Observational, cross-sectional, and retrospective study with an analytical approach, conducted between January 2023 and December 2024. A total of 596 eyes of premature newborns screened using the RetCam system and evaluated with artificial intelligence (VART) in seven NICUs in Barranquilla were included. Descriptive statistical analyses, inferential tests, and logistic regression were applied to identify factors associated with ROP. Results: The prevalence of ROP was 19.5%. The mean gestational age was 28.6 weeks, and the average birth weight was 1263.7 g. A significant association was found between lower gestational age and birth weight with the development of ROP (p < 0.001). Of the cases with ROP, 84.4% received treatment, with laser being the most commonly used modality (64.2%), followed by combinations with anti-VEGF (26.5%). AI used in the VART system demonstrated a sensitivity of 80% for ROP screening.
Introduction: Retinopathy of prematurity (ROP) is one of the leading causes of preventable childhood blindness, particularly in developing countries. In Barranquilla, where neonatal care has improved, there remains a need to update data on its incidence, as well as on the most commonly used diagnostic and therapeutic tools. In this context, artificial intelligence (AI) has been proposed as an alternative to improve timely screening and reduce the burden on specialists. Objective: To determine the clinical factors associated with the presence and severity of retinopathy of prematurity in neonates screened with artificial intelligence using VART and RetCam at a clinic in Barranquilla during the 2023–2024 period. Materials and Methods: Observational, cross-sectional, and retrospective study with an analytical approach, conducted between January 2023 and December 2024. A total of 596 eyes of premature newborns screened using the RetCam system and evaluated with artificial intelligence (VART) in seven NICUs in Barranquilla were included. Descriptive statistical analyses, inferential tests, and logistic regression were applied to identify factors associated with ROP. Results: The prevalence of ROP was 19.5%. The mean gestational age was 28.6 weeks, and the average birth weight was 1263.7 g. A significant association was found between lower gestational age and birth weight with the development of ROP (p < 0.001). Of the cases with ROP, 84.4% received treatment, with laser being the most commonly used modality (64.2%), followed by combinations with anti-VEGF (26.5%). AI used in the VART system demonstrated a sensitivity of 80% for ROP screening.
Descripción
Palabras clave
Retinopatía del prematuro, Prematuridad, Inteligencia artificial, Screening neonatal, Factores de riesgo, Anti-VEGF, Fotocoagulación láser