Automatic Detection of Microaneurysms in Fundus Images Using an Ensemble-Based Segmentation Method

In this study, a novel method for automatic microaneurysm detection in color fundus images is presented. The proposed method is based on three main steps: (1) image breakdown to smaller image patches, (2) inference to segmentation models, and (3) reconstruction of the predicted segmentation map from output patches. The proposed segmentation method is based on an ensemble of three individual deep networks, such as U-Net, ResNet34-UNet and UNet++. The performance evaluation is based on the calculation of the Dice score and IoU values. The ensemble-based model achieved higher Dice score (0.95) and IoU (0.91) values compared to other network architectures. The proposed ensemble-based model demonstrates the high practical application potential for detection of early-stage diabetic retinopathy in color fundus images.

Automated summary

This study presents a novel method for automatically detecting microaneurysms in color fundus images. The proposed method consists of three main steps: image breakdown into smaller patches, inference to segmentation models, and reconstruction of the predicted segmentation map. The proposed segmentation method utilizes an ensemble of three different deep networks, including U-Net, ResNet34-UNet, and UNet++. The performance evaluation is based on Dice score and IoU values, and the ensemble-based model achieves higher scores compared to other network architectures. The proposed ensemble-based model demonstrates high potential for the practical application of early-stage diabetic retinopathy detection in color fundus images.