The Efficacy of Microaneurysms Detection With and Without Vessel Segmentation in Color Retinal Images

TL;DR

This study compares the effectiveness of microaneurysm detection in retinal images with and without vessel segmentation, demonstrating high sensitivity in both approaches and highlighting the importance of vessel segmentation in diagnostic algorithms.

Contribution

The paper introduces an automated method combining Radon transform and CNNs for microaneurysm detection, comparing detection performance with and without vessel segmentation.

Findings

With vessel segmentation, sensitivity >85% and 11 false positives per image.

Without vessel segmentation, sensitivity ~90% with 73 false positives per image.

Vessel segmentation improves specificity, aiding retinal pathology diagnosis.

Abstract

Computer-Aided Diagnosis systems are required to extract suitable information about retina and its changes. In particular, identifying objects of interest such as lesions and anatomical structures from the retinal images is a challenging and iterative process that is doable by image processing approaches. Microaneurysm (MAs) are one set of these changes caused by diabetic retinopathy (DR). In fact, MAs detection is the main step for the identification of DR in the retinal images analysis. The objective of this study is to apply an automated method for the detection of MAs and compare the results of detection with and without vessel segmentation and masking either in the normal or abnormal image. The steps for detection and segmentation are as follows. In the first step, we did preprocessing, by using top-hat transformation. Our main processing was included applying Radon transform, to segment the vessels and masking them. At last, we did the MAs detection step using a combination of Laplacian-of-Gaussian and Convolutional Neural Networks. To evaluate the accuracy of our proposed method, we compare the output of our proposed method with the ground truth that collected by ophthalmologists. With vessel segmentation, our algorithm found a sensitivity of more than 85% in the detection of MAs with 11 false-positive rates per image for 100 color images in a local retinal database and 20 images of a public dataset (DRIVE). Also without vessel segmentation, our automated algorithm finds a sensitivity of about 90% in the detection of MAs with 73 false positives per image for all 120 images of two databases. In conclusion, with vessel segmentation, we have acceptable sensitivity and specificity, as a necessary step in some diagnostic algorithm for retinal pathology.