Abstract:BACKGROUND AND OBJECTIVE:Retinal microaneurysm (MA) is one of the earliest clinical signs of diabetic retinopathy(DR). Its detection is essential for controlling DR and preventing vision loss. However, the spatial scale of MA is extremely small and the contrast to surrounding background is subtle, which make MA detection challenging. The purpose of this work is to automatically detect MAs from fundus images.METHODS:Our MA detector involves two stages: MA candidate extraction and classification. In MA candidate extraction stage, local minimum region extraction and block filtering are used to exploit the regions where MA may exist. In this way, most of irrelavent background regions are discarded , which subsequently facilitates the training of MA classifier. In the second stage, multiple features are extracted to train the MA classifier. To distinguish MA from vascular regions, we propose a series of descriptors according to the cross-section profile of MA. Specially, as MAs are small and their contrast to surroundings is subtle, we propose local cross-section transformation (LCT) to amplify the difference between the MA and confusing structures. Finally, an under-sampling boosting-based classifier (RUSBoost) is trained to determine whether the candidate is an MA.RESULTS:The proposed method is evaluated on three public available databases i.e. e-ophtha-MA, DiaretDB1 and ROC training set. It achieves high sensitivities for low false positive rates on the three databases. Using the FROC metric, the final scores are 0.516, 0.402 and 0.293 respectively, which are comparable to existing state-of-the-art methods.CONCLUSIONS:The proposed local cross-section transformation enhances the discrimination of descriptors by amplifying difference between MAs and confusing structures, which facilitates the classification and improves the detection performances. With the powerful descriptors, our method achieves state-of-the-art performances on three public datasets consistently.

Detection of Microaneurysms Using Multi-Scale Correlation Coefficients

Hierarchical Detection of Red Lesions in Retinal Images by Multiscale Correlation Filtering.

Automatic Analysis of Microaneurysms Turnover to Diagnose the Progression of Diabetic Retinopathy.

New hierarchical approach for microaneurysms detection with matched filter and machine learning

Automatic Microaneurysm Detection of Diabetic Retinopathy in Fundus Images

Automatic Microaneurysm Detection in Fundus Image Based on Local Cross-Section Transformation and Multi-Feature Fusion

A Magnified Adaptive Feature Pyramid Network for automatic microaneurysms detection

Retinal Microaneurysms Detection using Local Convergence Index Features

A Fast Localization and Extraction of Microaneurysm for Early Detection of Diabetic Retionopathy

A Novel Diabetic Retinopathy Detection Approach Based on Deep Symmetric Convolutional Neural Network

Retinopathy Online Challenge: Automatic Detection of Microaneurysms in Digital Color Fundus Photographs

Local Structure Awareness-Based Retinal Microaneurysm Detection with Multi-Feature Combination

Detecting Retinal Microaneurysms and Hemorrhages with Robustness to the Presence of Blood Vessels

Computer aided diagnosis of diabetic retinopathy based on multi-view joint learning

A Novel Lesion Detection Algorithm Based on Multi-Scale Input Convolutional Neural Network Model for Diabetic Retinopathy

Automatic detection of microaneurysms using DeTraC deep convolutional neural network classifier with woodpecker mating algorithm

Microaneurysms Detection in Retinal Fundus Images Based on Shape Constraint with Region-Context Features

An Automatic Detection Model of Microaneurysms Based on Improved FC-Densenet

Automatic Diabetic Retinopathy Screening Via Cascaded Framework Based on Image- and Lesion-Level Features Fusion

SAFP-Net: A spatially augmented feature pyramid network for automatic detection of microaneurysms

Computer Aided Diagnosis For Diabetic Retinopathy Based On Fundus Image