Ali Erginay · enciclopedia

Publicaciones › Cribado, telemedicina (OPHDIAT) y epidemiología de la retinopatía diabética

A multiple-instance learning framework for diabetic retinopathy screening

Un marco de aprendizaje de múltiples instancias para el cribado de la retinopatía diabética

Año2012
FuenteMedical image analysis
Tipoarticle · Journal Article, Research Support, N.I.H., Extramural, Research Support, Non-U.S. Gov't, Research Support, U.S. Gov't, Non-P.H.S.
Autores (8)Gwénolé Quellec, Mathieu Lamard, Michael D Abràmoff, Etienne Decencière, Bruno Lay, Ali Erginay, Béatrice Cochener, Guy Cazuguel
Posición de Erginay6 de 8
Citas105 (OpenAlex)
EnlacesDOI 10.1016/j.media.2012.06.003 · PubMed 22850462 · OpenAlex · Texto completo en acceso abierto
LíneasInteligencia artificial y análisis automático de imágenes Cribado, telemedicina (OPHDIAT) y epidemiología de la retinopatía diabética Retinopatía diabética: clínica, función y progresión

En pocas palabras

Se presentó un nuevo marco de aprendizaje automático para clasificar imágenes de retina. El sistema se entrenó con imágenes marcadas por clínicos y logró detectar patrones relevantes sin necesidad de segmentaciones manuales. Se aplicó a dos conjuntos de datos de imágenes de retina y obtuvo un alto rendimiento en la detección de lesiones de retinopatía diabética.

Resumen divulgativo generado automáticamente a partir del resumen original; puede contener imprecisiones.

Resumen original

A novel multiple-instance learning framework, for automated image classification, is presented in this paper. Given reference images marked by clinicians as relevant or irrelevant, the image classifier is trained to detect patterns, of arbitrary size, that only appear in relevant images. After training, similar patterns are sought in new images in order to classify them as either relevant or irrelevant images. Therefore, no manual segmentations are required. As a consequence, large image datasets are available for training. The proposed framework was applied to diabetic retinopathy screening in 2-D retinal image datasets: Messidor (1200 images) and e-ophtha, a dataset of 25,702 examination records from the Ophdiat screening network (107,799 images). In this application, an image (or an examination record) is relevant if the patient should be referred to an ophthalmologist. Trained on one half of Messidor, the classifier achieved high performance on the other half of Messidor (A(z)=0.881) and on e-ophtha (A(z)=0.761). We observed, in a subset of 273 manually segmented images from e-ophtha, that all eight types of diabetic retinopathy lesions are detected.

Términos MeSH

Algorithms Artificial Intelligence Diabetic Retinopathy Humans Image Enhancement Image Interpretation, Computer-Assisted Mass Screening Pattern Recognition, Automated Reproducibility of Results Retinoscopy Sensitivity and Specificity