Microaneurysms Research Articles

Evaluation of microaneurysms as predictors of therapeutic response to anti-VEGF therapy in patients with DME.

Administration of intravitreal anti-vascular endothelial growth factor (anti-VEGF) therapy is the first-line therapy for diabetic macular oedema (DME). However, some patients show no or insufficient response to repeated anti-VEGF injections. Therefore, it is necessary to identify factors that can predict this resistance against anti-VEGF treatment. Presence of microaneurysms (MAs) is a predictor of the development and progression of DME, but its relationship with the treatment response to the anti-VEGF agents is not well known. Therefore, we aimed to elucidate the relationship between the distribution of MAs and the response to anti-VEGF therapy in patients with DME. The number of MAs was measured before anti-VEGF therapy in each region using fluorescein angiography, indocyanine green angiography (IA), and optical coherence tomography angiography. Patients with DME were divided into the responder and non-responder groups after three loading phases. Differences in the distribution of MAs between the groups were investigated. Pre-treatment IA revealed more MAs in the nasal area in the non-responder group than in the responder group (10.7 ± 10.7 and 5.7 ± 5.7, respectively, in the nasal macula) (1.4 ± 2.1 and 0.4 ± 0.7, respectively, in the nasal fovea). Whereas, pre-treatment FA and OCTA could not reveal significantly difference between the groups. Detection of MAs in the nasal macula using pre-treatment IA may indicate resistance to anti-VEGF therapy. We recommend the clinicians confirm the presence of MAs in the nasal macula, as shown by IA, as a predictor of therapeutic response to anti-VEGF therapy in patients with treatment naive DME.

Comparison of 24×20 mm2 swept-source OCTA and fluorescein angiography for the evaluation of lesions in diabetic retinopathy

To compare ultra-widefield (24×20 mm2) swept-source optical coherence tomography angiography (SS-OCTA) and fluorescein angiography (FA) in the evaluation of diabetic retinopathy (DR) lesions. Forty-six eyes of 23 patients with treatment-naïve DR were included at Peking University People's Hospital from September 1, 2021, until December 31, 2021, as well as 23 age and gender matched healthy controls. Quantitative assessments of DR lesions on FA and SS-OCTA (superficial capillary plexus, SCP, 24×20 mm2) were performed. Area of fovea avascular zone (FAZ) was larger in DR cases than controls (0.34±0.069 mm2 vs 0.287±0.108 mm2, P=0.006). In DR eyes, the mean FAZ area was 0.34±0.069 and 0.334±0.087 mm2 on SS-OCTA and FA, respectively (P=0.428), while the median FAZ perimeter was 2.382 (IQR, 2.201-2.59) and 2.333 (IQR, 2.138-2.6) mm on SS-OCTA and FA images (P=0.733). There was no significant difference in the size of the non-perfusion area (NPA) between the images on SS-OCTA and FA (12.389, IQR 4.96-28.3 and 11.125, IQR 5-28.31 mm2, P=0.197). The median total microaneurysm (MA) count was 35 (IQR, 19-46) and 73 (IQR, 43-93) on SS-OCTA and FA (P<0.001), respectively. No significant difference in intra-retinal microvascular abnormality (IRMA) and neovascularization (NV) count was found between the two techniques. The intraclass coefficient (ICCs) of all the parameters above indicated stable repeatability. Ultra-widefield SS-OCTA represents a reliable, noninvasive, and quantitative imaging technique in the assessment of microvasculature in DR, which offers a potential substitute for FA in DR evaluation.

Combining transfer learning with retinal lesion features for accurate detection of diabetic retinopathy.

Diabetic retinopathy (DR) is a late microvascular complication of Diabetes Mellitus (DM) that could lead to permanent blindness in patients, without early detection. Although adequate management of DM via regular eye examination can preserve vision in in 98% of the DR cases, DR screening and diagnoses based on clinical lesion features devised by expert clinicians; are costly, time-consuming and not sufficiently accurate. This raises the requirements for Artificial Intelligent (AI) systems which can accurately detect DR automatically and thus preventing DR before affecting vision. Hence, such systems can help clinician experts in certain cases and aid ophthalmologists in rapid diagnoses. To address such requirements, several approaches have been proposed in the literature that use Machine Learning (ML) and Deep Learning (DL) techniques to develop such systems. However, these approaches ignore the highly valuable clinical lesion features that could contribute significantly to the accurate detection of DR. Therefore, in this study we introduce a framework called DR-detector that employs the Extreme Gradient Boosting (XGBoost) ML model trained via the combination of the features extracted by the pretrained convolutional neural networks commonly known as transfer learning (TL) models and the clinical retinal lesion features for accurate detection of DR. The retinal lesion features are extracted via image segmentation technique using the UNET DL model and captures exudates (EXs), microaneurysms (MAs), and hemorrhages (HEMs) that are relevant lesions for DR detection. The feature combination approach implemented in DR-detector has been applied to two common TL models in the literature namely VGG-16 and ResNet-50. We trained the DR-detector model using a training dataset comprising of 1,840 color fundus images collected from e-ophtha, retinal lesions and APTOS 2019 Kaggle datasets of which 920 images are healthy. To validate the DR-detector model, we test the model on external dataset that consists of 81 healthy images collected from High-Resolution Fundus (HRF) dataset and MESSIDOR-2 datasets and 81 images with DR signs collected from Indian Diabetic Retinopathy Image Dataset (IDRID) dataset annotated for DR by expert. The experimental results show that the DR-detector model achieves a testing accuracy of 100% in detecting DR after training it with the combination of ResNet-50 and lesion features and 99.38% accuracy after training it with the combination of VGG-16 and lesion features. More importantly, the results also show a higher contribution of specific lesion features toward the performance of the DR-detector model. For instance, using only the hemorrhages feature to train the model, our model achieves an accuracy of 99.38 in detecting DR, which is higher than the accuracy when training the model with the combination of all lesion features (89%) and equal to the accuracy when training the model with the combination of all lesions and VGG-16 features together. This highlights the possibility of using only the clinical features, such as lesions that are clinically interpretable, to build the next generation of robust artificial intelligence (AI) systems with great clinical interpretability for DR detection. The code of the DR-detector framework is available on GitHub at https://github.com/Janga-Lab/DR-detector and can be readily employed for detecting DR from retinal image datasets.

Comparison of widefield swept-source optical coherence tomography angiography with ultra-widefield fluorescein angiography for the evaluation of lesions in retinal vein occlusion

BackgroundTo compare widefield swept-source optical coherence tomography angiography (SS-OCTA) with ultra-widefield fundus fluorescein angiography (UWF-FA) for detecting retinal vein occlusion (RVO) lesions.MethodsThirty-four eyes of 32 patients with treatment-naïve RVO were enrolled at Peking University People’s Hospital from September 2021 to March 2022. Patients were imaged with a UWF-FA (200°) and a widefield SS-OCTA using 24 × 20 mm scan single capture. Quantitative assessments of RVO lesions such as foveal avascular zone (FAZ) area and perimeter, non-perfusion areas (NPA), number of microaneurysms (MAs), capillary changes and collateral vessels were performed.ResultsThe measurement of FAZ area and perimeter were comparable between SS-OCTA and UWF-FA (0.373 (range, 0.277–0.48) mm2 vs. 0.370 (range, 0.277–0.48) mm2, P = 0.818 and 2.480 (range, 2.011–2.998) vs. 2.330 (range, 2.027–2.807) mm, P = 0.536, respectively). Intraclass correlation coefficients (ICCs) of FAZ area and perimeter between SS-OCTA and UWF-FA was high (0.999, [0.997–0.999] and 0.996 [0.991–0.996], respectively), suggesting good agreement. The mean NPA area was larger on SS-OCTA than that on UWF-FA (89.977 ± 78.805 mm2vs. 87.944 ± 77.444 mm2, P = 0.037). The ICC of NPA area was also high (0.999, [0.999–1.000]). The median of total MA count was less on SS-OCTA than on UWF-FA (7 (range, 0–19) vs.12 (range, 0–23), P < 0.001). Agreement in detecting MAs between SS-OCTA and UWF-FA was found to be good (ICC = 0.920, [0.555–0.974]).The total capillary changes and collateral vessels count were less on UWF-FA than SS-OCTA (11 ± 9 vs 6 ± 7, P < 0.001 and 4 (range, 0–6) vs 0 (range, 0–0), P < 0.001, respectively). Agreement in detecting capillary changes and collateral vessels between OCTA and UWF-FA was found to be fair (ICC = 0.733, [0.081–0.905] and 0.564, [0.039–0.805], respectively).ConclusionCompared with UWF-FA, widefield SS-OCTA was found comparable or even superior in detecting FAZ, NPA, capillary changes and collateral vessels except MAs in RVO. Widefield SS-OCTA may offer a more efficient alternative to FA for diagnosis and monitoring RVO.

Deep attentive convolutional neural network for automatic grading of imbalanced diabetic retinopathy in retinal fundus images.

Automated fine-grained diabetic retinopathy (DR) grading was of great significance for assisting ophthalmologists in monitoring DR and designing tailored treatments for patients. Nevertheless, it is a challenging task as a result of high intra-class variations, high inter-class similarities, small lesions, and imbalanced data distributions. The pivotal factor for the success in fine-grained DR grading is to discern more subtle associated lesion features, such as microaneurysms (MA), Hemorrhages (HM), soft exudates (SE), and hard exudates (HE). In this paper, we constructed a simple yet effective deep attentive convolutional neural network (DACNN) for DR grading and lesion discovery with only image-wise supervision. Designed as a top-down architecture, our model incorporated stochastic atrous spatial pyramid pooling (sASPP), global attention mechanism (GAM), category attention mechanism (CAM), and learnable connected module (LCM) to better extract lesion-related features and maximize the DR grading performance. To be concrete, we devised sASPP combining randomness with atrous spatial pyramid pooling (ASPP) to accommodate the various scales of the lesions and struggle against the co-adaptation of multiple atrous convolutions. Then, GAM was introduced to extract class-agnostic global attention feature details, whilst CAM was explored for seeking class-specific distinctive region-level lesion feature information and regarding each DR severity grade in an equal way, which tackled the problem of imbalance DR data distributions. Further, the LCM was designed to automatically and adaptively search the optimal connections among layers for better extracting detailed small lesion feature representations. The proposed approach obtained high accuracy of 88.0% and kappa score of 88.6% for multi-class DR grading task on the EyePACS dataset, respectively, while 98.5% AUC, 93.8% accuracy, 87.9% kappa, 90.7% recall, 94.6% precision, and 92.6% F1-score for referral and non-referral classification on the Messidor dataset. Extensive experimental results on three challenging benchmarks demonstrated that the proposed approach achieved competitive performance in DR grading and lesion discovery using retinal fundus images compared with existing cutting-edge methods, and had good generalization capacity for unseen DR datasets. These promising results highlighted its potential as an efficient and reliable tool to assist ophthalmologists in large-scale DR screening.

DIABETIC RETINOPATHY LESION DETECTION FROM MULTISPECTRAL RETINAL IMAGES THROUGH NEURAL NETWORK

Diabetic Retinopathy (DR) is one of the fastest growing dysfunctions of human retina. Significant research has been conducted using RGB fundus imaging for automatic detection of retinal lesions affected by DR. However, due to only three imaging bands, the accuracy from RGB fundus images is unlikely to improve any further. In contrast to RGB imaging, multispectral imaging has the key advantage of multiple narrow wavelength bands that can be used as spectral features to improve the detection accuracy. Nevertheless, the inter and intra-retinal variation of color, contrast, and illumination is a challenge to process the multispectral images. In this study, a complete framework is proposed to develop and evaluate methods for automatic detection of DR lesions. A multispectral retinal image database, DIARETSPECDB1, is investigated in order to detect the most common DRs such as Microaneurysms (MA), Hard Exudates (HE) and Hemorrhages (HEM). The reflectance values of the spectral bands are used as features of a three-layer basic neural network (NN) to determine the baseline performance of multispectral data instead of any advanced model. According to the results, the model outperforms existing technique producing overall accuracy 94.5%, and the obtained specificity/sensitivity is 0.95/0.89, 0.97/0.89, and 0.88/0.84 for MAs, HEs and HEMs, respectively.

Concurrent Learning of Patch and Image-Based Features by CNN for Diabetic Retinopathy Detection

Diabetic Retinopathy (DR) is a disorder of the eyes allied with prolonged hyperglycemia and is the leading cause of blindness globally. The earliest signs of DR include the presence of microaneurysms (MA). This research proposes a paradigm that includes patch-level lesion classification and DR severity assessment. A framework is presented that utilizes a multi-task learning approach centered on a single data source and is capable of predicting two outcomes. The proposed method for patch-level MA categorization consists of only three stages and is based on the combination of auto feature extraction by deep learning and classification by the machine learning algorithm. After preprocessing, for feature descriptor learning, GoogLeNet is employed, followed by SVM for classification. A comprehensive approach is provided to use the patch and image-level feature vectors to grade DR severity. To assess the suggested technique for patch-level classification, an experiment was undertaken by using two widely known public datasets, namely E-ophtha and IDRiD datasets. The developed system achieved 98.67% accuracy, a sensitivity of 99.2%, specificity of 99.06% on e-ophtha-MA database, while a trained model is validated on the IDRiD database, it achieves an accuracy of 70.6%, the sensitivity of 67.62%, specificity of 73.58%. The MESSIDOR dataset is used to evaluate the performance of DR grading as well. The proposed system achieves an accuracy of 89.39%, a sensitivity of 91.1%, specificity of 79.31% for two-stage grade DR severity.

Automatic detection of microaneurysms in optical coherence tomography images of retina using convolutional neural networks and transfer learning

Microaneurysms (MAs) are pathognomonic signs that help clinicians to detect diabetic retinopathy (DR) in the early stages. Automatic detection of MA in retinal images is an active area of research due to its application in screening processes for DR which is one of the main reasons of blindness amongst the working-age population. The focus of these works is on the automatic detection of MAs in en face retinal images like fundus color and Fluorescein Angiography (FA). On the other hand, detection of MAs from Optical Coherence Tomography (OCT) images has 2 main advantages: first, OCT is a non-invasive imaging technique that does not require injection, therefore is safer. Secondly, because of the proven application of OCT in detection of Age-Related Macular Degeneration, Diabetic Macular Edema, and normal cases, thanks to detecting MAs in OCT, extensive information is obtained by using this imaging technique. In this research, the concentration is on the diagnosis of MAs using deep learning in the OCT images which represent in-depth structure of retinal layers. To this end, OCT B-scans should be divided into strips and MA patterns should be searched in the resulted strips. Since we need a dataset comprising OCT image strips with suitable labels and such large labelled datasets are not yet available, we have created it. For this purpose, an exact registration method is utilized to align OCT images with FA photographs. Then, with the help of corresponding FA images, OCT image strips are created from OCT B-scans in four labels, namely MA, normal, abnormal, and vessel. Once the dataset of image strips is prepared, a stacked generalization (stacking) ensemble of four fine-tuned, pre-trained convolutional neural networks is trained to classify the strips of OCT images into the mentioned classes. FA images are used once to create OCT strips for training process and they are no longer needed for subsequent steps. Once the stacking ensemble model is obtained, it will be used to classify the OCT strips in the test process. The results demonstrate that the proposed framework classifies overall OCT image strips and OCT strips containing MAs with accuracy scores of 0.982 and 0.987, respectively.

AOSLO-net: A Deep Learning-Based Method for Automatic Segmentation of Retinal Microaneurysms From Adaptive Optics Scanning Laser Ophthalmoscopy Images.

PurposeAccurate segmentation of microaneurysms (MAs) from adaptive optics scanning laser ophthalmoscopy (AOSLO) images is crucial for identifying MA morphologies and assessing the hemodynamics inside the MAs. Herein, we introduce AOSLO-net to perform automatic MA segmentation from AOSLO images of diabetic retinas.MethodAOSLO-net is composed of a deep neural network based on UNet with a pretrained EfficientNet as the encoder. We have designed customized preprocessing and postprocessing policies for AOSLO images, including generation of multichannel images, de-noising, contrast enhancement, ensemble and union of model predictions, to optimize the MA segmentation. AOSLO-net is trained and tested using 87 MAs imaged from 28 eyes of 20 subjects with varying severity of diabetic retinopathy (DR), which is the largest available AOSLO dataset for MA detection. To avoid the overfitting in the model training process, we augment the training data by flipping, rotating, scaling the original image to increase the diversity of data available for model training.ResultsThe validity of the model is demonstrated by the good agreement between the predictions of AOSLO-net and the MA masks generated by ophthalmologists and skillful trainees on 87 patient-specific MA images. Our results show that AOSLO-net outperforms the state-of-the-art segmentation model (nnUNet) both in accuracy (e.g., intersection over union and Dice scores), as well as computational cost.ConclusionsWe demonstrate that AOSLO-net provides high-quality of MA segmentation from AOSLO images that enables correct MA morphological classification.Translational RelevanceAs the first attempt to automatically segment retinal MAs from AOSLO images, AOSLO-net could facilitate the pathological study of DR and help ophthalmologists make disease prognoses.

Automatic Detection of Microaneurysms in OCT Images Using Bag of Features.

Diabetic retinopathy (DR) caused by diabetes occurs as a result of changes in the retinal vessels and causes visual impairment. Microaneurysms (MAs) are the early clinical signs of DR, whose timely diagnosis can help detecting DR in the early stages of its development. It has been observed that MAs are more common in the inner retinal layers compared to the outer retinal layers in eyes suffering from DR. Optical coherence tomography (OCT) is a noninvasive imaging technique that provides a cross-sectional view of the retina, and it has been used in recent years to diagnose many eye diseases. As a result, this paper attempts to identify areas with MA from normal areas of the retina using OCT images. This work is done using the dataset collected from FA and OCT images of 20 patients with DR. In this regard, firstly fluorescein angiography (FA) and OCT images were registered. Then, the MA and normal areas were separated, and the features of each of these areas were extracted using the Bag of Features (BOF) approach with the Speeded-Up Robust Feature (SURF) descriptor. Finally, the classification process was performed using a multilayer perceptron network. For each of the criteria of accuracy, sensitivity, specificity, and precision, the obtained results were 96.33%, 97.33%, 95.4%, and 95.28%, respectively. Utilizing OCT images to detect MAs automatically is a new idea, and the results obtained as preliminary research in this field are promising.

Regional Variety of Reduction in Retinal Thickness of Diabetic Macular Edema after Anti-VEGF Treatment.

Background and Objectives: The presence of refractory cases resistant to anti-vascular endothelial growth factor (VEGF) therapy for diabetic macular edema (DME) is a problem in clinical practice. This study aimed to explore the less responsive area of optical coherence tomography (OCT) 3D map the characteristics of naïve DME cases after their first anti-VEGF. Materials and Methods: In 46 patients with DME who received an intravitreal injection of anti-VEGF agents, retinal thickness in 100 sections of the macular area was measured by 3D-mapping mode using OCT before and 1 month after injection. The density of the microaneurysm (MA) was calculated using merged images of the OCT map and fluorescein angiography. Results: One month after injection, the central retinal thickness significantly decreased (p < 0.0001). In severe edema (retinal thickness more than 500 µm), the area percentages with a reduction rate of the retinal thickness greater than 30% and less than 5% were 6.4 ± 6.6% and 10.1 ± 4.6%, respectively. The reduction rate of the retinal thickness varied from section to section. The mutual distance between the areas of maximum thickness before and after the injection averaged 1.22 ± 0.62 mm apart. The reduction rate of retinal thickness in the thickest region before injection was significantly higher (p = 0.02), and that in the thickest region after injection was lower (p = 0.001) than in the other regions. MA density in the residual edema was significantly higher than in the edema-absorbed area (p = 0.03). Conclusion: DME has areas that show low response to the reduction in retinal thickness with anti-VEGF therapy. A high density of MA may be associated with this pathogenesis.

AI-Based Automatic Detection and Classification of Diabetic Retinopathy Using U-Net and Deep Learning

Artificial intelligence is widely applied to automate Diabetic retinopathy diagnosis. Diabetes-related retinal vascular disease is one of the world’s most common leading causes of blindness and vision impairment. Therefore, automated DR detection systems would greatly benefit the early screening and treatment of DR and prevent vision loss caused by it. Researchers have proposed several systems to detect abnormalities in retinal images in the past few years. However, Diabetic Retinopathy automatic detection methods have traditionally been based on hand-crafted feature extraction from the retinal images and using a classifier to obtain the final classification. DNN (Deep neural networks) have made several changes in the previous few years to assist overcome the problem mentioned above. We suggested a two-stage novel approach for automated DR classification in this research. Due to the low fraction of positive instances in the asymmetric Optic Disk (OD) and blood vessels (BV) detection system, preprocessing and data augmentation techniques are used to enhance the image quality and quantity. The first step uses two independent U-Net models for OD (optic disc) and BV (blood vessel) segmentation. In the second stage, the symmetric hybrid CNN-SVD model was created after preprocessing to extract and choose the most discriminant features following OD and BV extraction using Inception-V3 based on transfer learning, and detects DR by recognizing retinal biomarkers such as MA (microaneurysms), HM (hemorrhages), and exudates (EX). On EyePACS-1, Messidor-2, and DIARETDB0, the proposed methodology demonstrated state-of-the-art performance, with an average accuracy of 97.92%, 94.59%, and 93.52%, respectively. Extensive testing and comparisons with baseline approaches indicate the efficacy of the suggested methodology.

Diabetic Retinopathy Detection from Fundus Images of the Eye Using Hybrid Deep Learning Features.

Diabetic Retinopathy (DR) is a medical condition present in patients suffering from long-term diabetes. If a diagnosis is not carried out at an early stage, it can lead to vision impairment. High blood sugar in diabetic patients is the main source of DR. This affects the blood vessels within the retina. Manual detection of DR is a difficult task since it can affect the retina, causing structural changes such as Microaneurysms (MAs), Exudates (EXs), Hemorrhages (HMs), and extra blood vessel growth. In this work, a hybrid technique for the detection and classification of Diabetic Retinopathy in fundus images of the eye is proposed. Transfer learning (TL) is used on pre-trained Convolutional Neural Network (CNN) models to extract features that are combined to generate a hybrid feature vector. This feature vector is passed on to various classifiers for binary and multiclass classification of fundus images. System performance is measured using various metrics and results are compared with recent approaches for DR detection. The proposed method provides significant performance improvement in DR detection for fundus images. For binary classification, the proposed modified method achieved the highest accuracy of 97.8% and 89.29% for multiclass classification.

Early Diabetic Retinopathy Detection Using Deep Learning Algorithms

Abstract: Diabetic Retinopathy (DR) is caused by the high blood glucose level which causes micro vascular complications in eyes which lead to vision loss. (MA)Microaneurysms formation in the retinal is sign of diabetic rent ropy which can be cured at early stage. Finding Microaneurysms (MA) presence in the eye image and recognition of diabetic retinopathy at early stage is difficult. Technology of deep learning makes it easier and efficient for analysis of eyes to detect MA presence which can be done by image detection and segmentation with good performance and accuracy using deep learning algorithms. This will help us to differentiated between affected retina and non-affected one. The given system can use deep convolution neural network for semantic segmentation of fundus images which increase efficiency and accuracy of NPDR (non- proliferated diabetic retinopathy) prediction. Keywords: PCA Principle Component Analysis DA Diabetic Retinopathy MA Microaneurysms NPDR Non proliferated diabetic retinopathy BPNN Back propagation neural network

Charles Jacques Bouchard (1837-1915) and the Charcot-Bouchard aneurysm.

Charles Jacques Bouchard was a distinguished French physician and scientist of the early 19th century. Despite his humble beginnings, Bouchard was able to achieve meteoric success within the scientific and medical fields, establishing himself as one of the most influential physician-scientists of his time. This was in part due to his superb commitment, as well as the prosperity engendered by the strong influence of his teachers, which can be seen as a testament to the importance of mentorship in medicine. Besides his myriad contributions, Bouchard is most well known for describing the Charcot-Bouchard aneurysm in 1866 alongside his mentor Jean-Martin Charcot, linking them for the first time to intracranial hemorrhage. Bouchard's thesis entitled "A Study of Some Points in the Pathology of Cerebral Hemorrhage" was regarded by some as the most original and important of all recent works on the subject of cerebral hemorrhage at the time of publication. Sadly, the great relationship Bouchard shared with his mentor Charcot would later deteriorate into perhaps one of the most well-known student-mentor quarrels in the history of medicine. Herein, the authors present a historical recollection of Bouchard's life, career, and contributions to medicine, as well as the famous controversy with Jean-Martin Charcot.

Diabetic retinopathy detection using deep learning based neural network along with machine learning algorithms

Diabetic retinopathy (DA) is an eye disease caused by retinal damage as a result of long-term diabetes mellitus. Microaneurysms (MA)are an indicator of DA and are small red spots formed on the retina caused by the ballooning of a weak blood artery.The DA is mainly classified between Proliferate diabetic retinopathy (PDR) and Non-proliferative diabetic retinopathy (NPDR). Non-proliferative is an earlier stage of DA. In our study we will classify the images into 5 stages based on their severity of DA taken from a dataset. The existing modelshave usedLogistic Regression (LR), Support Vector Machine (SVM), gradient boosting techniques such as XGBoost and Logistic Regression with Elastic-Net penalty (LR-EN), to classify wavelet features among the groups. In our project study we used theDeeplearning-based algorithmFast R-CNN (Region Convolutional Neural Network) to build the model and tested its accuracy in training as well as testing the same model with other Machine learning techniques like Decision Tree, k-nearest neighbors (k-NN) classifier, GaussianNaïve Bayes, Kernel-SVM. Our project study shows that Decision Tree had the best training accuracy with 99.31% whereas in case ofthe best predicted testing accuracy it is k-Nearest Neighbors (k-NN) Classifier with 71.29%.

Subarachnoid hemorrhage due to rupture of very small aneurysms of the anterior part of the circle of Willis

Objective: To assess the frequency of ruptures of very small cerebral aneurysms, features of the perioperative period and outcomes, in comparison with the rupture of ordinarily sized aneurysms.Material and methods: A comparative analysis of the group of patients with ruptured cerebral miliary aneurysms (n = 18) and the group of patients with ruptured cerebral aneurysms of regular size (n = 308) was carried out. All patients underwent open surgery in the first 3 days after the rupture (osteoplastic craniotomy, microsurgical aneurysm clipping). We compared gender, age of patients, severity of the patient’s condition at the moment of admission, severity of subarachnoid hemorrhage, location of aneurysms, aspect ratio, duration of the operation, frequency of intraoperative ruptures, postoperative mortality.Results: It was found that miliary aneurysm rupture occurs in 5.5% of all patients with cerebral aneurysm rupture. The most common cases of rupture of very small aneurysms were in women (77.7%), with a mean age of 50.8 years. Aneurysms of the anterior communicating artery (66.6%) with a narrow neck (average aspect ratio – 2.1) were the most common. Patients with rupture of very small aneurysms were 7.9% more likely to be admitted in a state of subcompensation or decompensation (Hunt-Hess IV–V), they had massive subarachnoid hemorrhage (Fisher III) 19.6% more often than with ruptured aneurysms of regular size. On average, operations in cases of very small aneurysms lasted 30 minutes less than clipping of ordinary aneurysms, but were complicated by intraoperative rupture twice as often (38.8% and 16.5%, respectively). Postoperative mortality in the group of patients with ruptured miliary aneurysms was 5.7% higher than in patients with ruptured aneurysms of regular size.Conclusion: Rupture of cerebral miliary aneurysms is relatively rare. Women of 50–60 years old with very small aneurysms of the anterior communicating artery with a narrow neck constitute the main group of such patients. Massive subarachnoid hemorrhage and severe condition of patients on admission are more common with miliary aneurysms than with ordinarily sized aneurysms. The small size of the aneurysm and the work near the rupture determine the more frequent contact intraoperative rupture when the neck is exposed as compared to operations on larger aneurysms, which negatively affects the treatment outcomes in this group of patients.

Detection of microaneurysms in color fundus images based on local Fourier transform

Retinal microaneurysm (MA) is one of the early clinical signs of diabetic retinopathy (DR), which is essential for the early diagnosis of DR. Its shape is often shown as a red dot, which makes detection of MA challenging because of its tiny size. There are many effective MA automatic detection methods having been proposed, but these methods are hard to describe the overall shape of Mas. This paper proposes an automatic MA detection method based on local Fourier transform which splits the original image into small independent images and calculates their local Fourier transforms in isolation. After that, the object is not the whole image but a series of independent small images. We also extract a statistical feature to check for normality and design a new feature to represent the difference among every direction. After that, a group of candidate points are extracted from the images. Finally, some features are extracted from the candidate points, and the confidence scores of the candidate points are obtained by using random forest algorithm for classification. The proposed method obtained a FROC curve score of 0.847 on the E-Ophtha MA database, which is better than most existing methods. In the ROC database, although the FROC curve score is only 0.283, the AUC can reach 0.961, which is better than most methods. It can be seen that the proposed method can achieve a better result with low noise and large sample size.

An Ensemble Classifier Based on Individual Features for Detecting Microaneurysms in Diabetic Retinopathy

Individuals with diabetes are more likely to develop Diabetic Retinopathy (DR), a chronic ailment that can lead to blindness if left undiagnosed. Early-stage Diabetic Retinopathy (DR) is characterized by Microaneurysms (MA), which appear as tiny red lesions on the retina. This paper investigates a unique approach for the automated early identification of microaneurysms in eye images. A unique ensemble classifier technique is suggested in this work. Classifiers like SVM, KNN, Decision Tree, and Naïve Bayes are chosen in this study for building an ensemble model. After preprocessing the image , certain common image characteristics such as shape and intensity features were retrieved from the candidate. The mean absolute difference of each feature is computed. Based on mean ranges that would give improved classification results, an expert classifier is chosen and trained. The outputs of the classifiers are integrated for each of the distinct characteristics, and the number of categories that have been most frequently repeated is utilized to reach a final decision. The process has been comprehensively validated using two available open datasets, like e-ophtha and DIARETDB1. On the e-ophtha and DIARETDB1 datasets, the ensemble model achieved an AUC of 0.928 and 0.873, Sensitivity of 90.7% and 85%, Specificity of 90% and 91% respectively.

Response to 1-Year Fixed-Regimen Bevacizumab Therapy in Treatment-Naïve DME Patients: Assessment by OCT Angiography.

Purpose To evaluate the effectiveness of intravitreal bevacizumab treatment in patients with diabetic macular edema (DME) by assessing retinal changes using optical coherence tomography angiography (OCT-A). Methods This prospective study was performed in patients with treatment-naïve DME. The eyes of patients were imaged using a swept-source OCT system with a scan area of 6 × 6 mm. The DME patients with a central macular thickness (CMT) of ≥300 μm received nine bevacizumab injections within 12 months. The demographic, systemic, and ocular parameters, including the best-corrected visual acuity (BCVA), CMT, microaneurysm (MA) count, and foveal avascular zone (FAZ) area in both superficial capillary plexus (SCP) and deep capillary plexus (DCP), as well as vessel density in SCP, were assessed in the patients. In addition, the response (good or poor) of the DME eyes to bevacizumab treatment and the final visual acuity (BCVA of 75 letters) were analyzed. Results Seventy-seven eyes of DME patients were subjected to the final analysis. Bevacizumab treatment reduced CMT from 425.06 μm (±77.15) to 350.25 μm (±82.04) and improved BCVA by about 8.61 letters (from 64.73 to 73.34) in the patients. The mean number of MAs in SCP decreased from 3.51 ± 2.07 to 2.31 ± 1.15 (p < 0.001) and in DCP from 17.12 ± 11.56 to 12.21 ± 6.99 (p < 0.001), whereas the area of FAZ increased in SCP from 328.22 ± 131.38 to 399.70 ± 156.98 (p < 0.001) and in DCP from 571.13 ± 396.01 to 665.89 ± 412.77 (p = 0.001). The final BCVA letter score and CMT were statistically significant in both poor and good responders, as well as in BCVA < 75 and BCVA ≥ 75 groups. Conclusion The fixed-regimen intravitreal bevacizumab therapy was effective in treating DME. Apart from noninvasive visualization of microvascular damage, OCT-A showed limited usefulness in predicting treatment response. Although the study showed that the number of MAs was significantly reduced during treatment, which is an OCT-A predictor of a good response to bevacizumab treatment at a 12-month visit, commonly observed artifacts may reduce the usefulness of OCT-A.