Diabetic retinopathy grading based on multi-scale residual network and cross-attention module

Artificial intelligence using deep learning to screen for referable and vision-threatening diabetic retinopathy in Africa: a clinical validation study

Using Deep Learning Models to Characterize Major Retinal Features on Color Fundus Photographs

Facial expression methods play a vital role in human-computer interaction and other fields, but there are factors such as occlusion, illumination, and pose changes in wild facial recognition, as well as category imbalances between different datasets, that result in large variations in recognition rates and low accuracy rates for different categories of facial expression datasets. This study introduces RCL-Net, a method of recognizing wild facial expressions that is based on an attention mechanism and LBP feature fusion. The structure consists of two main branches, namely the ResNet-CBAM residual attention branch and the local binary feature (LBP) extraction branch (RCL-Net). First, by merging the residual network and hybrid attention mechanism, the residual attention network is presented to emphasize the local detail feature information of facial expressions; the significant characteristics of facial expressions are retrieved from both channel and spatial dimensions to build the residual attention classification model. Second, we present a locally improved residual network attention model. LBP features are introduced into the facial expression feature extraction stage in order to extract texture information on expression photographs in order to emphasize facial feature information and enhance the recognition accuracy of the model. Lastly, experimental validation is performed using the FER2013, FERPLUS, CK+, and RAF-DB datasets, and the experimental results demonstrate that the proposed method has superior generalization capability and robustness in the laboratory-controlled environment and field environment compared to the most recent experimental methods.

Artificial intelligence for diabetic retinopathy screening in Africa

The main goal of this paper is to evaluate the performance of deep learning with Residual Attention Network (RAN) for brain tumour classification. Digitalised Magnetic Resonance Image (MRI) datasets obtained from Malaysian hospitals and other sources are utilised in this paper. The MRI datasets consist of information of those patients who are 20 years old and above, both male and female. The RAN algorithm is trained and tested using the MRI datasets. The algorithm performance is evaluated based on training accuracy, testing accuracy, validation accuracy, and validation loss metrices. Moreover, a comparative analysis is done with Residual Neural Network (ResNet) and Convolutional Neural Network (CNN) using the same datasets. The findings from this study prove that RAN provides the best performance among the three algorithms. ResNet has good performance, with an accuracy ranging from 67% to 87%. The standard CNN algorithm does not perform well, with a very inconsistent accuracy of between 57% and 71%. RAN produces the highest and most consistent accuracy, which is 94% and above. Further explanation is provided in this paper to prove the efficiency of RAN for the classification of brain tumours.

Diabetic Retinopathy (DR) is eye disease caused through the retinal damage induced through long-term diabetes mellitus illness. In the initial stage, DR has minor vision difficulties or no symptoms but it outputs in vision loss if it is not treated early. However, overfitting occurs when a model learns noise or specific features from a small dataset which leading to poor generalization. To solve this issue, this research proposed an Otsu Multi-level Adaptive Thresholding with Deep Convolution Neural Network (OMAT+DCNN) approach for DR segmentation and classification. The Contrast Limited Adaptive Histogram Equalization (CLAHE) is used for preprocessing input image which enhances the contrast level and image visibility without amplifying noise. The OMAT is used for segmenting the affected regions and ResNet50 is used for extracting Deep Learning (DL) features. Then, relevant features are selected through Multi-Verse Optimizer (MVO) and DCNN is used for classification. The OMAT improves segmentation accuracy by considering multiple thresholds which adopts to local intensity variations within the image, even when the dataset is limited. DCNN capturing intricate patterns and spatial hierarchies within images and its deep architecture allows to learn complex representations which leads to enhance accuracy. The parameters like dice coefficient, Intersection over Union (IoU), Mean IoU (MIoU), precision, accuracy, f1-score and sensitivity are used for assessing performance. The OMAT+DCNN reaches 0.993 of accuracy when compared to existing techniques such as Convolutional Transformer Network (CTNet) and Residual Attention Network (RAN).

High-resolution retinal fundus images are critical for diagnosing diabetic retinopathy, yet clinical datasets often contain low-resolution images that obscure fine vascular structures essential for accurate diagnosis. Existing super-resolution methods face a fundamental trade-off: convolutional neural networks produce overly smooth results, while generative adversarial networks (GANs) risk creating hallucinated artifacts. We propose the Dual-Transformer Residual Super-Resolution Network (DTRSRN), a hybrid architecture combining Swin Transformers for global context modeling with a parallel residual Convolutional Neural Network (CNN) pathway for fine-grained vascular detail preservation. Our key innovation lies in using Fractal Dimension analysis to quantitatively measure retinal vascular morphology preservation. Experimental results on three benchmark datasets demonstrate that DTRSRN achieves 33.64 dB PSNR for 2 × super-resolution, out performing state-of-the-art methods including SwinIR (+0.96 dB), HAT (+0.37 dB), and ResShift (+0.30 dB). Critically, DTRSRN achieves 17.0% improvement in vascular structure preservation (ΔD f = 0.0987) compared to the best baseline, demonstrating superior clinical relevance for retinal image enhancement.

Retinal diseases should be detected early and accurately to avoid vision loss. It is a research paper on RetinoScope, a state-of-the-art AI-based multi-fusion platform to perform automated classification of retinal diseases such as diabetic retinopathy, glaucoma, and macular degeneration based on fundus images as well as OCT images. RetinoScope is a triple-stream model, combining a hybrid Convolutional Neural Network (CNN) to extract local features, a Residual Attention Network (RANet) to enhance the features with lesion focus, and a Vision Transformer (ViT) to have a global understanding. Cross-Attention Mechanism (CAM) is used to combine features of these streams at the feature level and Weighted Probabilistic Averaging is used to combine features at the decision level with a Genetic Algorithm (GA-WPA). The framework utilizes preprocessing methods such as contrast enhancement, filtering of vessels and optic discs and augmentation of the data in order to enhance generalization. Given explainable AIs such as Grad-CAM and Attention Rollout, explainable heatmaps are clinically interpretable. RetinoScope was evaluated on multiple datasets, which are APTOS 2019, MESSIDOR 2, Retina OCT 2017, DRIONSDB and HRF and showed 98.9% accuracy, 99.1% sensitivity, and 98.7 % specificity, which is higher than single architecture models. These findings indicate that RetinoScope provides a powerful, interpretable, and clinical solution to automated retinal disease diagnosis, which allows cross-modality synergy and allows further development of teleophthalmology solutions.

Handwritten Chinese recognition which involves variant writing style, thousands of character categories and monotonous data mark process is a long-term focus in the field of pattern recognition research. The existing methods are facing huge challenges including the complex structure of character/line-touching, the discriminate ability of similar characters and the labeling of training datasets. To deal with these challenges, an end-to-end residual attention handwritten Chinese paragraph text recognition method is proposed, which uses fully convolutional neural networks as the main structure of feature extraction and employs connectionist temporal classification as a loss function. The novel residual attention gate block is more helpful in extracting essential features and making the training of deep convolutional neural networks more effective. In addition, we introduce the operations of batch bilinear interpolation which implement the mapping of two dimension text representation to one dimension text line representation without any position information of characters or text lines, and greatly reduce the labeling workload in preparing training datasets. In experimental, the proposed method is verified with two widely adopted handwritten Chinese text datasets, and achieves competitive results to the current state-of-the-art methods. Without using any position information of characters and text line, an accuracy rate of 90.53% is obtained in CASIA-HWDB test set.

Diabetic retinopathy (DR) is a complication of prolonged diabetic mellitus causing damage to the micro-vascular system in retina. Computer-aided detection of DR can significantly reduce time, effort, and cost in early diagnosis and thereby restrict further complications. In this paper, we present a deep learning based framework for automated DR stage classification from fundus retinal images.We design and develop a custom build Convolutional Neural Network (CNN) model named Model-IVS to extract meaningful and discriminative features from the retinal fundus images. The proposed hybrid-CNN combines the multi-scale pyramidal feature processing concept of inception model with the convolutional pipeline based feature extraction concept of VGG-Nets and adopts multi-scale feature concatenation based short-cut connections of DenseNet with learnable convolutions. For the binary classification tasks, we employ Model-IVS as a feature extractor and use Support Vector Machine (SVM) classifier on the reduced features obtained through principal component analysis. The proposed hybrid model is end-to-end trained incrementally on a class-balanced dataset curated from the train set of EyePACS dataset. The cross-dataset generalization of the proposed hybrid CNN based DR classification framework is evaluated on APTOS, MESSIDOR-1 and MESSIDOR-2 datasets. Experimental results demonstrate that our model has outperformed all the existing state-of-the-art methods for DR severity grading task on APTOS dataset. The proposed Model-IVS has achieved best accuracy, AUC-ROC, and kappa score of 84.33%, 0.978 and 0.899, respectively in DR severity grading task. In binary classification tasks like DR screening and referable DR predictions, our Model-IVS feature extraction network with SVM classifier on PCA features has also achieved comparable performance results w.r.t. the state-of-the-art methods. In DR screening task, we recorded an accuracy of 97.84%, 87.42%, and 89.67% on APTOS, MESSIDOR-1, and MESSIDOR-2 datasets, respectively. In referable DR prediction, we achieved AUC-ROC score of 0.987, 0.928, and 0.873 on APTOS, MESSIDOR-1, and MESSIDOR-2 datasets, respectively.

Automatic defect detection is of great significance to the production process of modern industries, which will affect the product appearance and quality. And a great number of economic loss may be caused due to the defects in the industrial production process. Traditional manual defect inspection method is labor-intensive and time-consuming with subjective factors, and the detection performance is random and uncertain. Machine learning has shown a good detection ability on small-scale samples, but the detect detection task against with poor contrast, weak texture, etc., will affect the effective feature representation. To address the detection task of steel surface defects, combined with the strong context extraction ability of deep learning, an accurate deep defect classification network is proposed in the paper to provide an end-to-end detection scheme. Fused with the residual network (Resnet50) and spatial attention block, a residual attention network is proposed for effective feature representation, which could make the classification network better focus on the defect areas. Meanwhile, due to the scale information change among different defects, a multi-scale context fusion (MCF) block is proposed for effective multi-scale feature extraction, which is conducive for multi-scale object detection. Experimental results on public defect data set show that the proposed defect detection network could acquire a superior classification performance compared with some typical classification networks.

Accurate retinal vessel segmentation plays a vital role in the early detection and diagnosis of ophthalmologic dis- eases such as diabetic retinopathy, glaucoma, and hypertension. Traditional image processing and machine learning techniques often struggle to capture the fine vessel structures and complex variations present in retinal fundus images. To overcome these limitations, this study proposes a hybrid deep learning framework that combines Convolutional Neural Networks (CNNs) and Gen- erative Adversarial Networks (GANs) for precise retinal vessel segmentation. In the proposed method, the CNN component serves as the feature extractor, learning multi-scale spatial and contextual features from retinal images, while the GAN framework enhances segmentation accuracy by employing a discriminator network that distinguishes between real and generated vessel maps. This adversarial learning setup encourages the generator to produce vessel segmentations that closely resemble ground truth images. Additionally, preprocessing techniques such as contrast enhancement and noise reduction are applied to improve the visibility of retinal structures, and post-processing steps refine the vessel boundaries for higher accuracy. Experimental results on publicly available retinal image datasets, such as DRIVE and STARE, demonstrate that the pro- posed GAN-CNN model outperforms conventional CNN-based segmentation methods in terms of accuracy, sensitivity, specificity, and Dice coefficient. The system achieves robust performance even under varying illumination and noise conditions. These findings indicate that integrating adversarial learning with convolutional feature extraction can significantly enhance retinal vessel segmentation, paving the way for more efficient and automated retinal disease screening systems

1 - Classification of ocular diseases using transfer learning approaches and glaucoma severity grading

Diabetes mellitus, widely called diabetes, is a metabolic disorder that results in high glucose (blood sugar) level because of the inadequate production of the hormone called insulin. Diabetes, which affects the eyes and damages the retinal vessels, is known as diabetic retinopathy (DR). DR may even lead to blindness without prior symptoms. The existing handcrafted-based methods do not provide potential results in detecting retinal features and morphological differences in the retinal fundus images. Predominantly, in health care, deep learning makes a tremendous part in providing promising and more accurate results since it works well with unlabeled data and can easily learn low-level features from the training data. It acts as an assistant for the ophthalmologists in robustly examining retinal abnormalities. This chapter aims to discuss the state-of-the-art systems for detecting diabetic eye diseases with both traditional and deep learning techniques, and a statistical comparison is also made using various performance metrics. On comparing the accuracies and other metrics, the deep learning technique, convolutional neural network (CNN) solely surpasses the other conventional methods such as morphological methods, geometrical methods, pixel-based methods, and so on. Further, in this chapter, publicly available pertinent retinal image databases are also discussed.KeywordsConvolutional neural network (CNN)Deep learningDiabetes mellitusDiabetic retinopathy

Retinal fundus images play a crucial role in the early detection of eye problems, aiding in timely diagnosis and treatment to prevent vision loss or blindness. With advancements in technology, Convolutional Neural Network (CNN) algorithms have emerged as effective tools for recognition, delineation, and classification tasks. This study proposes a comprehensive review of CNN algorithms used for retinal fundus image segmentation and classification. Our review follows a systematic approach, exploring diverse repositories to identify studies employing CNN to segment and classify retinal fundus images. Utilizing CNNs in the segmentation and classification of retinal fundus images can enhance the precision of segmentation outcomes and alleviate the sole dependence on human experts. This approach enables more accurate segmentation results, reducing the burden on human experts. A total of sixty-two studies are included in our review, analyzing aspects such as database usage and the advantages and disadvantages of the methods employed. The review provides valuable insights, limitations, observations, and future directions in the field. Despite certain limitations, the findings indicate that CNN algorithms consistently achieve high accuracies. The comprehensive examination of the included studies sheds light on the potential of CNN in retinal fundus image analysis.