Retinal Imaging

Machine learning for ophthalmic image analysis

I will present part of my work of ophthalmic image analysis using machine/deep learning techniques as part of the closure event of the Deep Learning for Medical Image Analysis course that Enzo Ferrante is teaching at the University of Buenos Aires.

Towards a Glaucoma Risk Index Based on Simulated Hemodynamics from Fundus Images

We designed a method to summarize hemodynamic parameters obtained by 0D simulations so that they can be applied for glaucoma detection. We observed certain correlation between glaucoma and these hemodynamic features.

An ensemble deep learning based approach for red lesion detection in fundus images

We introduced a hybrid red lesion detection model based on a combination of deep learning based features and hand crafted descriptors.

Retinal blood vessel segmentation in high resolution fundus photographs using automated feature parameter estimation

We developed a simple linear regression model that is able to estimate the hyperparameters of a fully-connected CRF model for blood vessel segmentation in fundus images.

Proliferative diabetic retinopathy characterization based on fractal features: Evaluation on a publicly available dataset

We develop a fractal based model for detecting proliferative diabetic retinopathy cases from fundus pictures.

Retinal blood vessel segmentation in high resolution fundus photographs using automated parameter estimation

I will present my SIPAIM 2017 paper on blood vessel segmentation in retinal images using FC-CRF with feature parameters estimated using linear regression.

Aprendizaje automático para asistencia al diagnóstico de enfermedades visuales basado en imágenes de fondo de ojo (Machine learning for ophthalmic screening and diagnostics from fundus images)

My PhD thesis.

Convolutional neural network transfer for automated glaucoma identification

We use pretrained VGG-S and OverFeat architectures as feature extractors for glaucoma detection in fundus pictures. We were able to get almost 0.8 AUC without fine-tuning the networks!

Convolutional Neural Network Transfer for Automated Glaucoma Identification

I will present my SIPAIM 2016 paper on glaucoma classification using logistic regression and pre-trained neural networks as feature extractors.

A Discriminatively Trained Fully Connected Conditional Random Field Model for Blood Vessel Segmentation in Fundus Images

We present an extensive description and evaluation of our method for blood vessel segmentation in fundus images based on a discriminatively trained fully connected conditional random field model.