Efficient microaneurysm segmentation in retinal images via a lightweight Attention U-Net for early DR diagnosis.

Early stage symptom of the diabetic retinopathy is Microaneurysms (MAs). Diabetic retinopathy is graded with the help of number of MAs in fundus image. Detection of MAs in initial stage of diabetic retinopathy may prevent the vision loss. In this work a new approach is proposed for finding the MAs in fundus image. This method follows the three steps for detection of MAs. Enhancement of local contrast of the fundus image and removal of blood vessels are completed in first two steps. In the last step, MAs are detected based on the size and shape features. The proposed method is tested on DIARETDB1 database. The proposed method achieved sensitivity of 87.5% for recognizing MAs in fundus images.

A novel automated system of discriminating Microaneurysms in fundus images

ABSTRACTDiabetic retinopathy (DR) is a complication of diabetes that can lead to vision impairment and even permanent blindness. The increasing number of diabetic patients and a shortage of ophthalmologists highlight the need for automated screening tools for early detection. Microaneurysms (MAs) are the earliest indicators of DR. However, detecting MAs in fundus images is a challenging task due to its small size and subtle features. Additionally, low contrast, noise, and lighting variations in fundus images, such as glare and shadows, further complicate the detection process. To address these challenges, we incorporated image enhancement techniques such as green channel utilization, gamma correction, and median filtering to improve image quality. Furthermore, to enhance the performance of the MA detection model, we employed a lightweight feature pyramid network (FPN) with a pretrained ResNet34 backbone to capture multiscale features and the convolutional block attention module (CBAM) to enhance feature selection. CBAM applies spatial and channel‐wise attention, which allows the model to focus on the most relevant features for improved detection. We evaluated our method on the IDRID and E‐ophtha datasets, achieving a sensitivity of 0.607 and F1 score of 0.681 on IDRID and a sensitivity of 0.602 and F1 score of 0.650 on E‐ophtha. These experimental results show that our proposed method gives better results than previous methods.

Microaneurysms (MAs) are the earliest detectable diabetic retinopathy (DR) lesions. Thus, the ability to automatically detect MAs is critical for the early diagnosis of DR. However, achieving the accurate and reliable detection of MAs remains a significant challenge due to the size and complexity of retinal fundus images. Therefore, this paper presents a novel MA detection method based on a deep neural network with a multilayer attention mechanism for retinal fundus images. First, a series of equalization operations are performed to improve the quality of the fundus images. Then, based on the attention mechanism, multiple feature layers with obvious target features are fused to achieve preliminary MA detection. Finally, the spatial relationships between MAs and blood vessels are utilized to perform a secondary screening of the preliminary test results to obtain the final MA detection results. We evaluated the method on the IDRiD_VOC dataset, which was collected from the open IDRiD dataset. The results show that our method effectively improves the average accuracy and sensitivity of MA detection.

Vision loss happens due to diabetic retinopathy (DR) in severe stages. Thus, an automatic detection method applied to diagnose DR in an earlier phase may help medical doctors to make better decisions. DR is considered one of the main risks, leading to blindness. Computer-Aided Diagnosis systems play an essential role in detecting features in fundus images. Fundus images may include blood vessels, exudates, micro-aneurysm, hemorrhages, and neovascularization. In this paper, our model combines automatic detection for the diabetic retinopathy classification with localization methods depending on weakly-supervised learning. The model has four stages; in stage one, various preprocessing techniques are applied to smooth the data set. In stage two, the network had gotten deeply to the optic disk segment for eliminating any exudate's false prediction because the exudates had the same color pixel as the optic disk. In stage three, the network is fed through training data to classify each label. Finally, the layers of the convolution neural network are re-edited, and used to localize the impact of DR on the patient's eye. The framework tackles the matching technique between two essential concepts where the classification problem depends on the supervised learning method. While the localization problem was obtained by the weakly supervised method. An additional layer known as weakly supervised sensitive heat map (WSSH) was added to detect the ROI of the lesion at a test accuracy of 98.65%, while comparing with Class Activation Map that involved weakly supervised technology achieved 0.954. The main purpose is to learn a representation that collect the central localization of discriminative features in a retina image. CNN-WSSH model is able to highlight decisive features in a single forward pass for getting the best detection of lesions.

Diabetic retinopathy (DR) is a complication of diabetes that affects the eyes. It occurs when high blood sugar levels damage the blood vessels in the retina, the light-sensitive tissue at the back of the eye. Therefore, there is a need to detect DR in the early stages to reduce the risk of blindness. Transfer learning is a machine learning technique where a pre-trained model is used as a starting point for a new task. Transfer learning has been applied to diabetic retinopathy classification with promising results. Pre-trained models, such as convolutional neural networks (CNNs), can be fine-tuned on a new dataset of retinal images to classify diabetic retinopathy. This manuscript aims at developing an automated scheme for diagnosing and grading DR and HR. The retinal image classification has been performed using three phases that include preprocessing, segmentation and feature extraction techniques. The pre-processing methodology has been proposed for reducing the noise in retinal images. A-CLAHE, DNCNN and Wiener filter techniques have been applied for the enhancement of images. After pre-processing, blood vessel segmentation in retinal images has been performed utilizing OTSU thresholding and mathematical morphology. Feature extraction and classification have been performed using transfer learning models. The segmented images were then classified using Modified ResNet 101 architecture. The performance for enhanced images has been evaluated on PSNR and shows better results as compared to the existing literature. The network is trained on more than 6000 images from MESSIDOR and ODIR datasets and achieves the classification accuracy of 98.72%.

Simple methods for segmentation and measurement of diabetic retinopathy lesions in retinal fundus images

Timely detection and treatment of microaneurysms (MA) is a critical step to prevent the development of vision-threatening eye diseases such as diabetic retinopathy. However, detecting MAs in fundus images is a highly challenging task due to the large variation of imaging conditions. In this paper, we focus on developing an interleaved deep mining technique to cope intelligently with the unbalanced MA detection problem. Specifically, we present a clinical report guided multi-sieving convolutional neural network (MS-CNN) which leverages a small amount of supervised information in clinical reports to identify the potential MA regions via a text-to-image mapping in the feature space. These potential MA regions are then interleaved with the fundus image information for multi-sieving deep mining in a highly unbalanced classification problem. Critically, the clinical reports are employed to bridge the semantic gap between low-level image features and high-level diagnostic information. Extensive evaluations show our framework achieves 99.7% precision and 87.8% recall, comparing favorably with the state-of-the-art algorithms. Integration of expert domain knowledge and image information demonstrates the feasibility to reduce the training difficulty of the classifiers under extremely unbalanced data distribution.

Automatic detection of microaneurysms in fundus images based on multiple preprocessing fusion to extract features

Diabetic Retinopathy (DR) is vision threatening and can be prevented with early diagnosis and treatment. This can be achieved with the regular screening of patients known to have Diabetes for 5 years or more. Once detected with DR, it is important for doctors to maintain the progress of the disease down the line. This includes identification and marking of DR features in the fundus images. Manual marking of DR features like exudates and hemorrhages is tedious and error prone job for opthalmologists. Detection of DR is widely done with fundus imaging technique. To help aid ophthalmologists, the DR features in fundus images can be automatically marked using machine learning algorithms [1]. In this paper, a novel and generalized method for segmenting the fundus images is proposed. With our approach, retinal color images are partitioned into non-overlapping segments covering the entire image. Each segment, i.e., splat, contains pixels with similar color and spatial location. A novel method for automated generalised generation of splat is presented in this paper and further marking of the diseased splats is also proposed. The proposed method is tested on DIARETDB1 dataset, achieving accuracies of 87.7% for exudate patches, 84.6% for hemorrhage patches and 80.7% for normal patches.

One of the most vital components of diabetes mellitus that cause visual blindness is Diabetic Retinopathy (DR). The primary sign of DR in the retina is the presence of the microaneurysms (MAs) that cause because of injury in the retina as a long-term abnormality effect leads to diabetes mellitus. Early identification of the MAs helps us to reduce and prevent DR at the early stage. In this process, ophthalmologists continuously observe the previous and current fundus images obtained using the fundus camera manually which consumes more time and energy. In the proposed work, we describe a procedure for automatic detection of MAs by applying thresholding and mathematical morphology techniques, which alternates the tedious and time-consuming manual method. The set of techniques is used for detecting MAs from the RGB fundus image. Preprocess techniques are used to resize the input image. The exudates are eliminated using the thresholding technique. The optic disc (OD) and blood vessels are removed by implementing morphological methods, and the feature extraction is implemented, based on their size. The experiment was carried out on freely accessible dataset DIARETDB1. The performance of the presented method detects MAs from the RGB fundus image and the results obtained show us the positive sign of the proposed technique. This technique not only helps ophthalmologist in automatic detection of MAs, it also helps in keeping track of the current and previous results of a patient for diagnosis.

A method to assist in the diagnosis of early diabetic retinopathy: Image processing applied to detection of microaneurysms in fundus images

A lightweight CNN for Diabetic Retinopathy classification from fundus images

Diabetic retinopathy (DR) is the leading cause of blindness if not detected and treated in time and is a serious complication of diabetes. Since DR is a progressive eye disease, the early detection and diagnosis of DR is important to prevent patients from blindness. One of the most characteristic symptoms of DR is the presence of microaneurysm (MA) – the early sign of DR, which is hard to detect manually due to its small size. In this paper, we propose an automatic MA detection method based on region growing and region classification. We solve two problems: 1) given a fundus image, how to automatically partition the image into regions that may or may not contain MAs through a region growing approach, and 2) given a region in a fundus image, how to automatically evaluate whether this region contains MA by feeding the features of the region into an artificial neural network (ANN) for classification. The proposed approach involves image preprocessing, region growing, feature selection and classification steps. In the experiment, the public dataset DIAbetic RETinopathy DataBase 1 (DIARETDB1) is used to provide training/testing data and ground truth. The proposed method can achieve the performance with sensitivity 86.6%, specificity 96.3%, and accuracy 93.9%, for automatic MA detection.

Diabetic Retinopathy (DR) is one of the leading causes of blindness. The early detection and treatment of DR is significant to save the human vision. The presence of microaneurysms (MAs) is the first sign of the disease. The correct identification of MAs is an essential for finding of DR at the early stages. In this paper, we propose a three phase system for efficient recognition of MAs. The tentative MA lesions are recovered from the fundus image in the first stage. To accurately classify an extracted candidate region into MA or non-MA, the second stage prepares an attribute vector for each tentative MA lesion based on shape, intensity and statistical properties. The third stage is a classification step to classify as MAs and Non-MAs for early stage detection of DR. We present a holoentropy enabled decision tree classifier which combines entropy and total correlation. The best feature for decision tree is selected based on holoentropy to enhance the correctness of the classification. The implemented system is experimented using fundus image database DIARETDB1. The proposed method achieved an overall accuracy of 97.67%.The proposed system has detected the MAs with higher performance using simple features and holoentropy based decision tree classifier. The proposed system is suitable for early stage detection of DR through the accurate identification of MAs.