RetNet: Retinal Disease Detection using Convolutional Neural Network

In medical applications, ocular OCT (optical coherence tomography) is used to assess glaucoma, macular degeneration, diabetic macular edema and other eye diseases as it is capable of showing the cross-sections of tissue layers. The creation of new blood vessels in the choroid layer of the eye is known as choroidal neovascularization (CNV). The aging and macular degeneration will represent the symptom DRUSEN. Our sharp central vision is affected due to DRUSEN. An irreversible vision loss is caused in diabetic patients due to diabetic macular edema (DME). It is mainly due to the leaking of blood vessels in the retina. This research work focus on designing a clinical decision support system to assist the ophthalmologist in classifying the three different types of eye diseases. The existing nine pre-trained CNN models are used for this purpose. The extracted features are used to generate the trained model that is further used for eye disease classification. The training accuracy, validation accuracy, training loss and validation loss are computed for 100 iterations for each pre-trained CNN models during training and validation. The trained model obtained after training is used as input to the classifier, which classifies the images under-diagnosis into NORMAL (normal eye), CNV, DME and DRUSEN. The performance metrics of the classifier designed using each pre-trained models are evaluated and compared for four classes independently. The test results show that the performance of the classifier implemented using the pre-trained model InceptionV3 is better than all other models.

Artificial Neural network as a power tool is now working in more and more areas such as recognition of handwritten digits and faces, sentence classification and audio recognition. This project aims at using convolutional neural network (CNN) to recognize handwritten digits provided by MNIST datasets with TensorFlow tool. Several variables such as dropout rate, epoch, fully connected layer with different neuron amount and filter amount were investigated in the experiments to investigate the performance of the model. Both training and validation accuracy were detected and gave us a direct reflection of the influence of different settings. With appropriate 0.2 dropout rate and 50 epochs, the model can accomplish 99.81% training accuracy and 99.38% validation accuracy. By adding fully connected layers with reasonable neuron amount, both training and validation accuracy were improved, and this modification had no extra running time required in our work. The filter number in the convolutional layers was also an important factor in the model performance. More filters can extract more features from the raw images and thus increase the accuracy of 99.95% for training accuracy and 99.47% for validation accuracy. However, this modification was found to have a longer running time.

Diabetic retinopathy (DR) is a common eye disease that people get from diabetes. About 33.7% of the people with diabetes have DR. With our datas, which are pictures of the eyeball with and without DR, we tried different convolutional neural network (CNN) models to get the best accuracy score. We tested our datas with a default CNN model, and 5 different pre-trained models: MobileNet, VGG16, VGG19, Inception V3, and Inception ResNet V2. The default CNN model didn’t perform very well, getting only 10.4%. The pre-trained model also didn’t perform as good as expected, so we decided to use GRU with the models, which increases the score. For the higher accuracy, we added bidirectional GRU to train the whole parameters in the model. The 5 different pre-trained models scored an average of 74.2% accuracy score, and Inception ResNet V2 with bidirectional GRU included scored the highest accuracy, achieving 83.57%. For additional study, we used a class activation map to spot the abnormal parts of the eyeball with DR, and we could spot abnormal veins and bleeding on the eyeball. However, our research has limitations on that we did not use the segmentation methods, which is more advanced technique compared to classification, such as U-net, Fully Convolutional Network (FCN), Deep Lab V3, and Feature Pyramid Network. Furthermore, even though our model classified 5 different classes, the fact that the highest accuracy score was lower than 90% is also a limitation. For further study, we would prepare a masking method for applying segmentation methods to our dataset.

Background: The recent outbreak of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and the disease corresponding to it (coronavirus disease 2019; COVID-19) has been declared a pandemic by the World Health Organization. COVID-19 has become a global crisis, shattering health care systems, and weakening economies of most countries. The current methods of testing that are employed include reverse transcription polymerase chain reaction (RT-PCR), rapid antigen testing, and lateral flow testing with RT-PCR being used as the golden standard despite its accuracy being at a mere 63%. It is a manual process which is time consuming, taking about an average of 48 hours to obtain the results. Alternative methods employing deep learning techniques and radiologic images are up and coming. Methods: In this paper, we used a dataset consisting of COVID-19 and non-COVID-19 folders for both X-Ray and CT images which contained a total number of 17,599 images. This dataset has been used to compare 3 (non-pre-trained) CNN models and 5 pre-trained models and their performances in detecting COVID-19 under various parameters like validation accuracy, training accuracy, validation loss, training loss, prediction accuracy, sensitivity and the training time required, with CT and X-Ray images separately. Results: Xception provided the highest validation accuracy (88%) when trained with the dataset containing the X- ray images while VGG19 provided the highest validation accuracy (81.2%) when CT images are used for training. Conclusions: The model, VGG16, showed the most consistent performance, with a validation accuracy of 76.6% for CT images and 87.76% for X-ray images. When comparing the results between the modalities, models trained with the X-ray dataset showed better performances than the same models trained with CT images. Hence, it can be concluded that X-ray images provide a higher accuracy in detecting COVID-19 making it an effective method for detecting COVID-19 in real life.

Alternative Dose for Choroidal Melanoma Treated With an Iodine-125 Radioactive Plaque: A Single-Institution Retrospective Study

Handwritten digit recognition is an important problem in the field of machine learning and computer vision, which is used in many practical applications such as document digitization, banking systems, postal automation, and academic assessment. Automatic identification of handwritten digits becomes a challenging task due to their variety in writing style, shape and thickness. The paper designed, implemented and evaluated a deep learning model based on the Convolutional Neural Network (CNN) using TensorFlow and Keras to accurately classify handwritten scores.The study used the MNIST dataset, which includes 60,000 training samples and 10,000 test samples in the form of grayscale images of 28 × 28 size. Under data pre-processing, techniques like normalization, re-reasoning and label encoding were adopted. The proposed CNN model incorporated convolution layers, max-pooling layers, fully-connected layers, and Softmax output layers, which enabled effective acquisition of spatial characteristics of handwritten digits.It was evident from the experimental results that the model achieved high classification accuracy, the difference between training and validation accuracy was minimal and the reliability of the classification was confirmed by confusion matrix analysis. The findings of the study indicate that the CNN-based deep learning model provides an accurate, reliable and practical solution for identifying handwritten marks.

In the realm of computer vision, image classification is a critical issue with many applications, including multimedia content analysis, security and surveillance, and medical imaging. The accuracy of image classification algorithms has considerably increased with the development of deep learning. The discipline of image classification has dramatically benefited from the usage of pre-trained Convolutional Neural Network (CNN) models. In this study, we perform image classification on the Wang dataset composed of 1000 images separated into ten categories, using six different pre-trained CNN models. The research made use of pre-trained versions of VGG16, Densenet, Mobilenet, Inception V3, Resnet50, and Xception models. We assessed the model performances in terms of training and testing accuracy. Testing accuracies for ten unique batches of data were calculated and averaged out. Densenet outperformed state-of-the-art models like Xception by a small margin, mainly due to low data availability. The findings of this study show how pre-trained models have advanced the field of image classification and shed light on how well these models perform in diverse image classification tasks. This study can serve as a valuable reference for researchers and practitioners in the field of computer vision and deep learning and help inform their choices when selecting pre-trained models for image classification, depending on their needs, while also considering data availability and computational constraints.

This paper proposes an efficient and cost effective deep learning architecture to detect the diabetic retinopathy in real time. Diabetes is a leading root cause of eye disease in patients. It illuminates eye vessels, and releases blood form vessels. Early detection of diabetic retinopathy is useful to reduce the risk of blindness or any hazard. In this paper, after some preprocessing and data augmentation, Inception V3 is used as pre-trained model to extract the initial features set. Convolutional neural network has been used with attention layers. These additional CNN layers are added to extract the deep features to improve classification performance and accuracy. Initially, the model was proposed by Kevin Mader in Kaggle. The paper introduced additional layers in proposed model and improved the validation and testing accuracy significantly. More than 90% validation accuracy was achieved with the proposed Convolutional Neural Network model. Testing accuracy was improved up to 5%. This improvement in accuracy is very significant because the dataset is imbalanced and contains noisy images. It is concluded that global average pooling (GAP) based attention mechanism increased deep learning architecture accuracy to detect the Diabetic Retinopathy in imbalanced and noisy image dataset

Diabetes affects 40–45% of Diabetic Retinopathy (DR) patients in the US. Early detection of DR may prevent or postpone vision deterioration, but it is difficult since the disorder often manifests with few symptoms until it is too late to treat. Clinically, DR is routinely treated using fundus images, with an estimated 200 million cases worldwide and over 400,000 deaths each year. A great deal of progress has been made by applying machine learning algorithms to the fundus images. As a result, image classification and detection have become reliable techniques for detecting the severity of diabetic retinopathy. Convolutional Neural Networks (CNNs) play a crucial role in the image classification and detection process by capturing various images' details, enabling a fast and efficient method for detecting diabetic retinopathy. CNN pre-trained models such as ResNet50, Inception V3,and EfficientNetB7 have substantially improved their performance in the ImageNet Large-Scale Visual Recognition Competition. In addition, these pre-trained models are more precise and inexpensive to train because of the shorter connections between their input and output layers. This work proposes an approach for image classification that ensembles three pre-trained models, namely: EfficientNetB7, ResNet50V2, and Inception V3,to perform the classification of the diabetic retinopathy subtypes. Our proposed method achieves 97.43 % accuracy by adjusting the weights of the pre-trained models in detecting D R using the EyePacs Dataset.

The use of intravitreal corticosteroids in the management of macular oedema has recently gained widespread acceptance. New long-acting steroid preparations and methods of delivery have facilitated the use of these new modalities. This review describes the various types of macular oedema for which this therapeutic option is used and the results.

The incidence of diabetes in Mauritius is amongst the highest in the world. Diabetic retinopathy (DR), a complication resulting from the disease, can lead to blindness if not detected early. The aim of this work was to investigate the use of transfer learning and data augmentation for the classification of fundus images into five different stages of diabetic retinopathy. The five stages are No DR, Mild nonproliferative DR, Moderate nonproliferative DR, Severe nonproliferative DR and Proliferative. To this end, deep transfer learning and three pre-trained models, VGG16, ResNet50 and DenseNet169, were used to classify the APTOS dataset. The preliminary experiments resulted in low training and validation accuracies, and hence, the APTOS dataset was augmented while ensuring a balance between the five classes. This dataset was then used to train the three models, and the best three models were used to classify a blind Mauritian test datum. We found that the ResNet50 model produced the best results out of the three models and also achieved very good accuracies for the five classes. The classification of class-4 Mauritian fundus images, severe cases, produced some unexpected results, with some images being classified as mild, and therefore needs to be further investigated.

The application of Artificial Intelligence (AI) in diagnosing retinal diseases represents a significant advancement in ophthalmological research, with the potential to reshape future practices in the field. This study explores the extensive applications and emerging research frontiers of AI in retinal diseases. This study aims to uncover the developments and predict future directions of AI research in retinal disease over the past decade. This study analyzes AI utilization in retinal disease research through articles, using citation data sourced from the Web of Science (WOS) Core Collection database, covering the period from January 1, 2014, to December 31, 2023. A combination of WOS analyzer, CiteSpace 6.2 R4, and VOSviewer 1.6.19 was used for a bibliometric analysis focusing on citation frequency, collaborations, and keyword trends from an expert perspective. A total of 2,861 articles across 93 countries or regions were cataloged, with notable growth in article numbers since 2017. China leads with 926 articles, constituting 32% of the total. The United States has the highest h-index at 66, while England has the most significant network centrality at 0.24. Notably, the University of London is the leading institution with 99 articles and shares the highest h-index (25) with University College London. The National University of Singapore stands out for its central role with a score of 0.16. Research primarily spans ophthalmology and computer science, with "network," "transfer learning," and "convolutional neural networks" being prominent burst keywords from 2021 to 2023. China leads globally in article counts, while the United States has a significant research impact. The University of London and University College London have made significant contributions to the literature. Diabetic retinopathy is the retinal disease with the highest volume of research. AI applications have focused on developing algorithms for diagnosing retinal diseases and investigating abnormal physiological features of the eye. Future research should pivot toward more advanced diagnostic systems for ophthalmic diseases.

Early detection of vitreoretinal diseases (VRDs) is critical for preventing vision loss, and currently relies on the examination and interpretation of multimodal imaging techniques. Artificial intelligence (AI) is emerging as a powerful tool to detect abnormalities in vitreoretinal morphology and ideally detect changes at earlier stages to allow for intervention. This meta-analysis evaluates and summarizes the diagnostic performance of AI models in the detection of VRDs using retinal imaging systems. This study was registered in PROSPERO (CRD42023450207). A comprehensive electronic search of PubMed/MEDLINE, EMBASE, and Web of Science was conducted by three independent reviewers up to August 2023. Study validity was assessed using the QUADAS-2 tool, which evaluates risk of bias across four domains and applicability concerns across three domains. Eligible articles were categorized into nine VRD subgroups-age related macular degeneration, diabetic retinopathy, retinal vascular diseases, retinal dystrophies, Cystoid macular edema, vitreoretinal interface disorders, retinal detachment, Central serous chorioretinopathy, and myopic retinopathy-and included in the meta-analysis. Pooled estimates of accuracy (PEA), sensitivity (PESen), and specificity (PESpe) were calculated for all selected studies. A total of 195 studies were included in the final analysis, yielding an overall PEA of 95.76% (95% CI: 95.0-96.47), PESen of 91.94% (95% CI: 90.72-93.08) and PESpe of 96.09% (95% CI: 95.27-96.79). In the subgroup analysis, most AI models had a PEA > 90%, especially convolutional neural networks (CNN), followed by support vector machine (SVM) and random forest (RF). AI diagnostic tools, particularly CNNs, have demonstrated robust performance in VRDs detection. However, results from studies with limited generalizability should be applied cautiously in real-world settings. Further exploration of emerging models, such as large language models (LLMs), is recommended.

Early detection of retinal diseases, such as diabetic retinopathy, is crucial in preventing irreversible vision loss. This study presents an automated diagnostic system utilizing deep learning techniques, specifically Convolutional Neural Networks (CNN) and pre-trained models like Mobile Net and VGG16. These models analyze retinal fundus images to detect abnormalities, including microaneurysms and hemorrhages, which indicate the presence of retinal diseases. CNN facilitates efficient feature extraction, while MobileNet and VGG16 enhance disease classification accuracy. MobileNet, with its lightweight design, is optimized for real-time mobile applications, ensuring fast and efficient detection. In contrast, VGG16 offers higher precision but demands greater computational resources.The proposed system is trained and tested on publicly available datasets to ensure robustness across diverse retinal images. A comparative analysis of MobileNet and VGG16 is conducted, focusing on accuracy, sensitivity, and specificity in detecting retinal abnormalities. Experimental results highlight the system’s potential to assist healthcare professionals by automating the diagnostic process, enabling early detection and timely medical intervention. This approach reduces reliance on manual screening, offering a scalable and accessible solution for retinal disease diagnosis.

The increasing use of convolutional neural networks (CNNs) has brought about a significant transformation in numerous fields, such as image categorization and identification. In the development of a CNN model to classify images of sericea lespedeza [SL; Lespedeza cuneata (Dum-Cours) G. Don] from weed images, four architectures were explored: CNN model variant 1, CNN model variant 2, the Visual Geometry Group (VGG16) model, and ResNet50. CNN model variant 1 (batch normalization with adjusted dropout method) demonstrated 100% validation accuracy, while variant 2 (RMSprop optimization with adjusted learning rate) achieved 90.78% validation accuracy. Pre-trained models, like VGG16 and ResNet50, were also analyzed. In contrast, ResNet50’s steady learning pattern indicated the potential for better generalization. A detailed evaluation of these models revealed that variant 1 achieved a perfect score in precision, recall, and F1 score, indicating superior optimization and feature utilization. Variant 2 presented a balanced performance, with metrics between 86% and 93%. VGG16 mirrored the behavior of variant 2, both maintaining around 90% accuracy. In contrast, ResNet50’s results revealed a conservative approach for class 0 predictions. Overall, variant 1 stood out in performance, while both variant 2 and VGG16 showed balanced results. The reliability of CNN model variant 1 was highlighted by the significant accuracy percentages, suggesting potential for practical implementation in agriculture. In addition to the above, a smartphone application for the identification of SL in a field-based trial showed promising results with an accuracy of 98–99%. The conclusion from the above is that a CNN model with batch normalization has the potential to play a crucial role in the future in redefining and optimizing the management of undesirable vegetation.