Trainable COSFIRE filters for vessel delineation with application to retinal images

By George Azzopardi, Nicola Strisciuglio, Mario Vento, N. Petkov. Medical Image Anal., 2015

707 citations recorded by the source corpus.

Retinal imaging provides a non-invasive opportunity for the diagnosis of several medical pathologies. The automatic segmentation of the vessel tree is an important pre-processing step which facilitates subsequent automatic processes that contribute to such diagnosis. We introduce a novel method for the automatic segmentation of vessel trees in retinal fundus images. We propose a filter that selectively responds to vessels and that we call B-COSFIRE with B standing for bar which is an abstraction for a vessel. It is based on the existing COSFIRE (Combination Of Shifted Filter Responses) approach. A B-COSFIRE filter achieves orientation selectivity by computing the weighted geometric mean of the output of a pool of Difference-of-Gaussians filters, whose supports are aligned in a collinear manner. It achieves rotation invariance efficiently by simple shifting operations. The proposed filter is versatile as its selectivity is determined from any given vessel-like prototype pattern in an automatic configuration process. We configure two B-COSFIRE filters, namely symmetric and asymmetric, that are selective for bars and bar-endings, respectively. We achieve vessel segmentation by summing up the responses of the two rotation-invariant B-COSFIRE filters followed by thresholding. The results that we achieve on three publicly available data sets (DRIVE: Se=0.7655, Sp=0.9704; STARE: Se=0.7716, Sp=0.9701; CHASE_DB1: Se=0.7585, Sp=0.9587) are higher than many of the state-of-the-art methods. The proposed segmentation approach is also very efficient with a time complexity that is significantly lower than existing methods.

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

View Paper: https://www.semanticscholar.org/paper/e69e6803ebd7fb411ade87db02453cbdfd417ba8

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

A Discriminatively Trained Fully Connected Conditional Random Field Model for Blood Vessel Segmentation in Fundus Images (2017, methodology): “The MCC is a correlation coefﬁcient between the manual and predicted binary segmentations, and it has been previously used for the evaluation of retinal vessel segmentation methods [11], [15], [17].”
A Cross-Modality Learning Approach for Vessel Segmentation in Retinal Images (2016, methodology): “Vessel segmentation is a basic step required for the quantitative analysis of retinal fundus images [3].”
Automated Vessel Segmentation Using Infinite Perimeter Active Contour Model with Hybrid Region Information with Application to Retinal Images (2015, methodology): “However, it is not clear to us why the COSFIRE method [11] has a higher AUC value than ours while its Se, Sp and Acc are all similar to ours on the DRIVE dataset, and all lower than ours on the STARE dataset.”
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].”
CE-Net: Context Encoder Network for 2D Medical Image Segmentation (2019, methodology): “We compare the proposed CE-Net with the state-of-the-art algorithms [5], [8], [7].”
Segmenting Retinal Blood Vessels With Deep Neural Networks (2016, methodology): “In [14], vessels are detected using a combination of co-linearly aligned difference-of-Gaussian filters, configured according to prototype patterns (e.”