ignaciorlando/tutorial-autoencoders-pytorch

Stars
2
Language
Jupyter Notebook
Created about 5 years ago
Updated about 5 years ago

ignaciorlando/tutorial-autoencoders-pytorch

ignaciorlando

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

red-lesion-detection

This code implements a red lesion detection method based on a combination of hand-crafted features and CNN based descriptors. Our paper is under revision now, so please do not use this repository until we release the paper.

fundus-vessel-segmentation-tbme

In this work, 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. Standard segmentation priors such as a Potts model or total variation usually fail when dealing with thin and elongated structures. We overcome this difficulty by using a conditional random field model with more expressive potentials, taking advantage of recent results enabling inference of fully connected models almost in real-time. Parameters of the method are learned automatically using a structured output support vector machine, a supervised technique widely used for structured prediction in a number of machine learning applications. Our method, trained with state of the art features, is evaluated both quantitatively and qualitatively on four publicly available data sets: DRIVE, STARE, CHASEDB1 and HRF. Additionally, a quantitative comparison with respect to other strategies is included. The experimental results show that this approach outperforms other techniques when evaluated in terms of sensitivity, F1-score, G-mean and Matthews correlation coefficient. Additionally, it was observed that the fully connected model is able to better distinguish the desired structures than the local neighborhood based approach. Results suggest that this method is suitable for the task of segmenting elongated structures, a feature that can be exploited to contribute with other medical and biological applications.

glaucoma-hemodynamics

This code corresponds to our MICCAI 2018 paper on retinal hemodynamics simulation. If you use this code, please cite: Orlando, JI, Barbosa Breda, J, van Keer, K, Blaschko, MB, Blanco, PJ and Bulant, C. "Towards a glaucoma risk index based on simulated hemodynamics from fundus images". MICCAI 2018.

fundus-fractal-analysis

This code corresponds to our Medical Physics paper with Karel van Keer, João Barbosa Breda, Hugo Luis Manterola, Matthew B. Blaschko and Alejandro Clausse, entitled "Proliferative Diabetic Retinopathy Characterization based on Fractal Features: Evaluation on a Publicly Available Data Set".

refuge-evaluation

This repository corresponds to the evaluation code used for the REFUGE challenge. Please, use it as a sanity check to verify that the format of your submissions are correct. The formatting instructions are provided in the website.

overfeat-glaucoma

This code corresponds to our paper with Matthew B. Blaschko, Elena Prokofyeva and Mariana del Fresno on Convolutional neural network transfer for automated glaucoma identification (SIPAIM 2016).

retinal-hemodynamics

Code for simulating the retinal hemodynamics from vessel segmentations extracted from fundus images.

cnn-dr-kaggle

Diabetic retinopathy detection using Convolutional Neural Networks. This code is part of our project with Pablo Rubí, Nicolás Dazeo, Carlos Bulant and Hugo Luis Manterola.

photoreceptor-segmentation

This code corresponds to the implementation of the U-shaped deep neural networks that we used in our Scientific Reports paper on photoreceptor segmentation in OCT scans

high-resolution-vessel-segmentation

This repository contains the blood vessel segmentation masks obtained using our method for blood vessel segmentation based on automated feature parameter estimation.

duia-ml-datasets

Jupyter Notebook

modelizacion-exactas

Este repositorio corresponde a nuestro trabajo de cursada/final de la materia Modelización, que se dicta como parte de la Licenciatura en Matemática de la Facultad de Ciencias Exactas de la UNCPBA. La materia también es válida como curso de postgrado para el Doctorado en Matemática Computacional e Industrial de la misma facultad. El trabajo consiste en optimizar mediante cutting-planes el recorrido más corto en un grafo con cotas de tiempo.