Detection and Location of Microaneurysms in Fundus Images Based on Improved YOLOv4

Abstract Microaneurysms (MA) are the initial symptoms of diabetic retinopathy (DR). Eliminating these lesions can effectively prevent DR at an early stage. However, due to the complex retinal structure and the different brightness and contrast of fundus images due to different factors, such as patients, environment, and acquisition equipment, it is difficult for existing detection algorithms to achieve accurate detection and location of the lesion. Therefore, the detection algorithm of improved YOLOv4(YOLOv4-Pro) was proposed. An improved Fuzzy C-Means (IFCM) clustering algorithm was proposed to optimize the anchor parameters of the target samples to improve the matching degree between the anchors and the feature graphs. The SENet attention module was then embedded in the backbone network to enhance the key information of the image and suppress the background information of the image to improve the confidence of MA. The spatial pyramid pooling (SPP) module was added to the neck to enhance the acceptance domain of the output characteristics of the backbone network to help separate important context information, and the model was verified on the Kaggle DR dataset and compared with other methods. The experimental results showed that, compared with other YOLOv4 network models with various structures, the improved YOLOv4 network model could significantly improve the automatic detection result. Compared with other network models and methods, the automatic detection accuracy of the improved YOLOv4 network model was better and an accurate position could be realized. Therefore, the proposed method of improved YOLOv4 performs better and can accurately and effectively detect and locate microaneurysms in fundus images.