RESUMEN
Purpose: To develop an automated method based on deep learning (DL) to classify macular edema (ME) from the evaluation of optical coherence tomography (OCT) scans. Methods: A total of 4230 images were obtained from data repositories of patients attended in an ophthalmology clinic in Colombia and two free open-access databases. They were annotated with four biomarkers (BMs) as intraretinal fluid, subretinal fluid, hyperreflective foci/tissue, and drusen. Then the scans were labeled as control or ocular disease among diabetic macular edema (DME), neovascular age-related macular degeneration (nAMD), and retinal vein occlusion (RVO) by two expert ophthalmologists. Our method was developed by following four consecutive phases: segmentation of BMs, the combination of BMs, feature extraction with convolutional neural networks to achieve binary classification for each disease, and, finally, multiclass classification of diseases and control images. Results: The accuracy of our model for nAMD was 97%, and for DME, RVO, and control were 94%, 93%, and 93%, respectively. Area under curve values were 0.99, 0.98, 0.96, and 0.97, respectively. The mean Cohen's kappa coefficient for the multiclass classification task was 0.84. Conclusions: The proposed DL model may identify OCT scans as normal and ME. In addition, it may classify its cause among three major exudative retinal diseases with high accuracy and reliability. Translational Relevance: Our DL approach can optimize the efficiency and timeliness of appropriate etiological diagnosis of ME, thus improving patient access and clinical decision making. It could be useful in places with a shortage of specialists and for readers that evaluate OCT scans remotely.
Asunto(s)
Aprendizaje Profundo , Retinopatía Diabética , Edema Macular , Oclusión de la Vena Retiniana , Retinopatía Diabética/diagnóstico , Retinopatía Diabética/diagnóstico por imagen , Humanos , Edema Macular/diagnóstico por imagen , Edema Macular/etiología , Reproducibilidad de los Resultados , Oclusión de la Vena Retiniana/diagnóstico , Oclusión de la Vena Retiniana/diagnóstico por imagen , Tomografía de Coherencia Óptica/métodosRESUMEN
Citomegalovirus, virus ADN bicatenario, de la familia Hepesviridae es capaz de ocasionar infecciones agudas o crónicas en el ser humano, invade a sus víctimas, aprovechándose principalmente de los menos inmunocompetentes quienes sufren mayor daño, especialmente los neonatos. Es de vital importancia su diagnóstico en la mujer de edad fértil, por su prevalencia y riesgo de transmisión vertical durante el embarazo, causando infección congénita que varía de 0.2 a 2.2% de todos los nacidos vivos, pudiendo ocasionar secuelas graves. El objetivo de esta investigación fue conocer la incidencia de infección congénita por Citomegalovirus en neonatos pretérmino y/o de bajo peso en el Hospital Central de Maracay en el período Marzo-Julio 2012, amplificando y detectando el ADN viral por reacción en cadena de polimerasa en muestras sanguíneas. Se estudiaron 106 neonatos menores de 1 semana de edad, siguiendo las normas bioéticas, se registraron sus antecedentes y se les tomó una muestra de sangre completa (1 mL) con anticoagulante. De las muestras procesadas se obtuvo 1 muestra positiva, mostrando incidencia de 0.9%. El neonato infectado, presentó sintomatología respiratoria producto de prematuréz, obtenido por cesárea a causa de ruptura prematura de membranas y procidencia de cordón. La incidencia de infección por Citomegalovirus en neonatos en el Hospital Central de Maracay, está dentro del rango encontrado en otros países, el paciente positivo se considera asintomático para Citomegalovirus tal como lo demuestra la literatura, un 90% de los casos nacen sin clínica evidente.
Citomegalovirus, double-stranded DNA virus, from the Herpesviridae family, can cause acute or chronic infections in humans, it invades his victims, mainly taking advantage of the less immunocompetent ones, suffering most damage, especially the newborn. It is vital to the diagnosis in women of childbearing age, because of its prevalence and the risk of vertical transmission during pregnancy, causing congenital infection ranging from 0.2 to 2.2% of all live births and can cause serious sequelae. The objective of this investigation was to know the incidence of citomegalovirus congenital infection in preterm newborns and / or low weight at the Maracay Central Hospital in the period March-July 2012, amplifying and detecting the viral DNA by polymerase chain reaction in blood. 106 newborns being less than one week old were studied, following bioethical standards, their background were registrated, and was taken a blood sample from them (1 mL). from the processed samples, one tested positive, showing an incidence of 0.9%. The infected newborn presented respiratory symptoms caused by the premature state, it was carried out by cesarean section caused by the rupture of the membranes and the prolapsed corde. The incidence of citomegalovirus infection in newborn at the Maracay Central Hospital, is within the range found in other countries, the positive case is considerate asymptomatic for citomegalovirus. as evidenced on text books, 90% of cases are born without clinical evidence.