An Ensemble Classification-Based Approach Applied to Retinal Blood Vessel Segmentation

By Muhammad Moazam Fraz, Paolo Remagnino, Andreas Hoppe, Bunyarit Uyyanonvara, Alicja R. Rudnicka, Christopher G. Owen, Sarah Ann Barman. IEEE Transactions on Biomedical Engineering, 2012

991 citations recorded by the source corpus.

This paper presents a new supervised method for segmentation of blood vessels in retinal photographs. This method uses an ensemble system of bagged and boosted decision trees and utilizes a feature vector based on the orientation analysis of gradient vector field, morphological transformation, line strength measures, and Gabor filter responses. The feature vector encodes information to handle the healthy as well as the pathological retinal image. The method is evaluated on the publicly available DRIVE and STARE databases, frequently used for this purpose and also on a new public retinal vessel reference dataset CHASE_DB1 which is a subset of retinal images of multiethnic children from the Child Heart and Health Study in England (CHASE) dataset. The performance of the ensemble system is evaluated in detail and the incurred accuracy, speed, robustness, and simplicity make the algorithm a suitable tool for automated retinal image analysis.

Subfields: Computer Vision and Pattern Recognition, Ophthalmology, Radiology, Nuclear Medicine and Imaging

View Paper: https://www.semanticscholar.org/paper/fc3c9b4f197bcacf3f8ca219cf3b29c5e49965d3

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Ophthalmology

In Radiology, Nuclear Medicine and Imaging

Citations in the corpus

An Ensemble of Fine-Tuned Convolutional Neural Networks for Medical Image Classification (2017, methodology): “Ensemble learning is a machine learning process in which better predictive performance is obtained by combining the results from multiple classification models into one high-quality classifier [12].”
Blood Vessel Segmentation of Fundus Images by Major Vessel Extraction and Subimage Classification (2015, methodology): “Each image is annotated by two independent human observers [10].”
Trainable COSFIRE filters for vessel delineation with application to retinal images (2015, background): “, 2011) report that they pre-process the images with background homogenization algorithms based on wide Gaussian kernels, while others (Soares et al., 2006; Fraz et al., 2012) explicitly state that they do not employ any pre-processing step.”
Learning with Limited Annotations: A Survey on Deep Semi-Supervised Learning for Medical Image Segmentation (2022, background): “Self-supervised tasks can also be trained jointly with target semi-supervised tasks [190],STARE[191], CHASE_DB1[192] Add a leaking GAN to pollute the discriminator by leaking information from the generator for more moderate generations”
Robust Retinal Vessel Segmentation via Locally Adaptive Derivative Frames in Orientation Scores (2016, methodology): “They can be generally divided into two main categories: supervised methods [2]–[5] and unsupervised methods [6]–[8].”
A Discriminatively Trained Fully Connected Conditional Random Field Model for Blood Vessel Segmentation in Fundus Images (2017, methodology): “Our experiments were carried out on DRIVE [44], STARE [46], CHASEDB1 [12], and HRF [24], [47], four standard publicly available datasets of fundus images used for the evaluation of blood vessel segmentation algorithms.”
A Three-Stage Deep Learning Model for Accurate Retinal Vessel Segmentation (2019, neutral): “[26] defined a”
A Cross-Modality Learning Approach for Vessel Segmentation in Retinal Images (2016, methodology): “The existence of vessel crossing, branching and centerline reflex makes it difficult [1], [8], [9] to segment the vessels accurately using artificially designed features.”
Joint Segment-Level and Pixel-Wise Losses for Deep Learning Based Retinal Vessel Segmentation (2018, methodology): “[22] used an ensemble system of bagged and boosted decision trees, where each feature vector was compromised of gradient, morphology, line strength and Gabor filter response.”
Hierarchical retinal blood vessel segmentation based on feature and ensemble learning (2015, methodology): “Fraz [6] 2012 0.7406 0.9807 0.9480 0.9747 17.”
DeepVessel: Retinal Vessel Segmentation via Deep Learning and Conditional Random Field (2016, methodology): “Note that there is no training set in the STARE and CHASE DB1 datasets, thus we only utilize the DRIVE training set to fine-tune the final parameters.”