Diabetic retinopathy detection through convolutional neural networks with synaptic metaplasticity

Identification of key biomarkers for early warning of diabetic retinopathy using BP neural network algorithm and hierarchical clustering analysis

Diabetic retinopathy is one of the most common microangiopathy in diabetes, essentially caused by abnormal blood glucose metabolism resulting from insufficient insulin secretion or reduced insulin activity. Epidemiological survey results show that about one third of diabetes patients have signs of diabetic retinopathy, and another third may suffer from serious retinopathy that threatens vision. However, the pathogenesis of diabetic retinopathy is still unclear, and there is no systematic method to detect the onset of the disease and effectively predict its occurrence. In this study, we used medical detection data from diabetic retinopathy patients to determine key biomarkers that induce disease onset through back propagation neural network algorithm and hierarchical clustering analysis, ultimately obtaining early warning signals of the disease. The key markers that induce diabetic retinopathy have been detected, which can also be used to explore the induction mechanism of disease occurrence and deliver strong warning signal before disease occurrence. We found that multiple clinical indicators that form key markers, such as glycated hemoglobin, serum uric acid, alanine aminotransferase are closely related to the occurrence of the disease. They respectively induced disease from the aspects of the individual lipid metabolism, cell oxidation reduction, bone metabolism and bone resorption and cell function of blood coagulation. The key markers that induce diabetic retinopathy complications do not act independently, but form a complete module to coordinate and work together before the onset of the disease, and transmit a strong warning signal. The key markers detected by this algorithm are more sensitive and effective in the early warning of disease. Hence, a new method related to key markers is proposed for the study of diabetic microvascular lesions. In clinical prediction and diagnosis, doctors can use key markers to give early warning of individual diseases and make early intervention.

Dual-Adaptive Heterojunction Synaptic Transistors for Efficient Machine Vision in Harsh Lighting Conditions

Photoadaptive synaptic devices enable in-sensor processing of complex illumination scenes, while second-order adaptive synaptic plasticity improves learning efficiency by modifying the learning rate in a given environment. The integration of above adaptations in one phototransistor device will provide opportunities for developing high-efficient machine vision system. Here, a dually adaptable organic heterojunction transistor as a working unit in the system, which facilitates precise contrast enhancement and improves convergence rate under harsh lighting conditions, is reported. The photoadaptive threshold sliding originates from the bidirectional photoconductivity caused by the light intensity-dependent photogating effect. Metaplasticity is successfully implemented owing to the combination of ambipolar behavior and charge trapping effect. By utilizing the transistor array in a machine vision system, the details and edges can be highlighted in the 0.4% low-contrast images, and a high recognition accuracy of 93.8% with a significantly promoted convergence rate by about 5 times are also achieved. These results open a strategy to fully implement metaplasticity in optoelectronic devices and suggest their vision processing applications in complex lighting scenes.

Vision transformer with masked autoencoders for referable diabetic retinopathy classification based on large-size retina image

Computer-aided diagnosis systems based on deep learning algorithms have shown potential applications in rapid diagnosis of diabetic retinopathy (DR). Due to the superior performance of Transformer over convolutional neural networks (CNN) on natural images, we attempted to develop a new model to classify referable DR based on a limited number of large-size retinal images by using Transformer. Vision Transformer (ViT) with Masked Autoencoders (MAE) was applied in this study to improve the classification performance of referable DR. We collected over 100,000 publicly fundus retinal images larger than 224×224, and then pre-trained ViT on these retinal images using MAE. The pre-trained ViT was applied to classify referable DR, the performance was also compared with that of ViT pre-trained using ImageNet. The improvement in model classification performance by pre-training with over 100,000 retinal images using MAE is superior to that pre-trained with ImageNet. The accuracy, area under curve (AUC), highest sensitivity and highest specificity of the present model are 93.42%, 0.9853, 0.973 and 0.9539, respectively. This study shows that MAE can provide more flexibility to the input image and substantially reduce the number of images required. Meanwhile, the pretraining dataset scale in this study is much smaller than ImageNet, and the pre-trained weights from ImageNet are not required also.

Competitive Perceptrons: The Relevance of Modeling New Bioinspired Properties Such as Intrinsic Plasticity, Metaplasticity, and Lateral Inhibition of Rate-Coding Artificial Neurons

This article supports the relevance of modeling new bioinspired properties in rate-coding artificial neurons, focusing on fundamental neural properties rarely implemented thus far in artificial neurons, such as intrinsic plasticity, the metaplasticity of synaptic strength, and the lateral inhibition of neighborhood neurons. All these properties are bioinspired through empirical models developed by neurologists, and this in turn contributes to taking perceptrons to a higher potential level. Metaplasticity and intrinsic plasticity are different levels of plasticity and are believed by neurologists to have fundamental roles in memory and learning and therefore in the performance of neurons. Assuming that information about stimuli is contained in the firing rate of the connections among biological neurons, several models of artificial implementation have been tested. Analyzing their results and comparing them with learning and performance of state-of-the-art models, relevant advances are made in the context of the developing Industrial Revolution 4.0 based on advances in Machine Learning, and they may even initiate a new generation of artificial neural networks. As an example, a single-layer perceptron that includes the proposed advances is successfully trained to perform the XOR function, called the Competitive Perceptron, which is a new bioinspired artificial neuronal model with the potential of non-linear separability, continuous learning, and scalability, which is suitable to build efficient Deep Networks, overcoming the basic limitations of traditional perceptrons that have challenged scientists for half a century.

Fine-grained attention & knowledge-based collaborative network for diabetic retinopathy grading

Accurate diabetic retinopathy (DR) grading is crucial for making the proper treatment plan to reduce the damage caused by vision loss. This task is challenging due to the fact that the DR related lesions are often small and subtle in visual differences and intra-class variations. Moreover, relationships between the lesions and the DR levels are complicated. Although many deep learning (DL) DR grading systems have been developed with some success, there are still rooms for grading accuracy improvement. A common issue is that not much medical knowledge was used in these DL DR grading systems. As a result, the grading results are not properly interpreted by ophthalmologists, thus hinder the potential for practical applications. This paper proposes a novel fine-grained attention & knowledge-based collaborative network (FA+KC-Net) to address this concern. The fine-grained attention network dynamically divides the extracted feature maps into smaller patches and effectively captures small image features that are meaningful in the sense of its training from large amount of retinopathy fundus images. The knowledge-based collaborative network extracts a-priori medical knowledge features, i.e., lesions such as the microaneurysms (MAs), soft exudates (SEs), hard exudates (EXs), and hemorrhages (HEs). Finally, decision rules are developed to fuse the DR grading results from the fine-grained network and the knowledge-based collaborative network to make the final grading. Extensive experiments are carried out on four widely-used datasets, the DDR, Messidor, APTOS, and EyePACS to evaluate the efficacy of our method and compare with other state-of-the-art (SOTA) DL models. Simulation results show that proposed FA+KC-Net is accurate and stable, achieves the best performances on the DDR, Messidor, and APTOS datasets.

Development and Internal Validation of Risk Assessment Models for Chronic Obstructive Pulmonary Disease in Coal Workers

Coal workers are more likely to develop chronic obstructive pulmonary disease due to exposure to occupational hazards such as dust. In this study, a risk scoring system is constructed according to the optimal model to provide feasible suggestions for the prevention of chronic obstructive pulmonary disease in coal workers. Using 3955 coal workers who participated in occupational health check-ups at Gequan mine and Dongpang mine of Hebei Jizhong Energy from July 2018 to August 2018 as the study subjects, random forest, logistic regression, and convolutional neural network models are established, and model performance is evaluated to select the optimal model, and finally a risk scoring system is constructed according to the optimal model to achieve model visualization. The training set results show that the logistic, random forest, and CNN models have sensitivities of 78.55%, 86.89%, and 77.18%; specificities of 85.23%, 92.32%, and 87.61%; accuracies of 81.21%, 85.40%, and 83.02%; Brier scores of 0.14, 0.10, and 0.14; and AUCs of 0.76, 0.88, and 0.78, respectively, and similar results are obtained for the test set and validation set, with the random forest model outperforming the other two models. The risk scoring system constructed according to the importance ranking of random forest predictor variables has an AUC of 0.842; the evaluation results of the risk scoring system shows that its accuracy rate is 83.7% and the AUC is 0.827, and the established risk scoring system has good discriminatory ability. The random forest model outperforms the CNN and logistic regression models. The chronic obstructive pulmonary disease risk scoring system constructed based on the random forest model has good discriminatory power.

A Prospective Study on Diabetic Retinopathy Detection Based on Modify Convolutional Neural Network Using Fundus Images at Sindh Institute of Ophthalmology & Visual Sciences

Diabetic Retinopathy (DR) is the most common complication that arises due to diabetes, and it affects the retina. It is the leading cause of blindness globally, and early detection can protect patients from losing sight. However, the early detection of Diabetic Retinopathy is an difficult task that needs clinical experts' interpretation of fundus images. In this study, a deep learning model was trained and validated on a private dataset and tested in real time at the Sindh Institute of Ophthalmology & Visual Sciences (SIOVS). The intelligent model evaluated the quality of the test images. The implemented model classified the test images into DR-Positive and DR-Negative ones. Furthermore, the results were reviewed by clinical experts to assess the model's performance. A total number of 398 patients, including 232 male and 166 female patients, were screened for five weeks. The model achieves 93.72% accuracy, 97.30% sensitivity, and 92.90% specificity on the test data as labelled by clinical experts on Diabetic Retinopathy.

A Deep Learning-Based Framework for Retinal Disease Classification

This study addresses the problem of the automatic detection of disease states of the retina. In order to solve the abovementioned problem, this study develops an artificially intelligent model. The model is based on a customized 19-layer deep convolutional neural network called VGG-19 architecture. The model (VGG-19 architecture) is empowered by transfer learning. The model is designed so that it can learn from a large set of images taken with optical coherence tomography (OCT) and classify them into four conditions of the retina: (1) choroidal neovascularization, (2) drusen, (3) diabetic macular edema, and (4) normal form. The training datasets (taken from publicly available sources) consist of 84,568 instances of OCT retinal images. The datasets exhibit all four classes of retinal disease mentioned above. The proposed model achieved a 99.17% classification accuracy with 0.995 specificities and 0.99 sensitivity, making it better than the existing models. In addition, the proper statistical evaluation is done on the predictions using such performance measures as (1) area under the receiver operating characteristic curve, (2) Cohen's kappa parameter, and (3) confusion matrix. Experimental results show that the proposed VGG-19 architecture coupled with transfer learning is an effective technique for automatically detecting the disease state of a retina.

Enhancing diabetic retinopathy classification using deep learning

Prolonged hyperglycemia can cause diabetic retinopathy (DR), which is a major contributor to blindness. Numerous incidences of DR may be avoided if it were identified and addressed promptly. Throughout recent years, many deep learning (DL)-based algorithms have been proposed to facilitate psychometric testing. Utilizing DL model that encompassed four scenarios, DR and its stages were identified in this study using retinal scans from the "Asia Pacific Tele-Ophthalmology Society (APTOS) 2019 Blindness Detection" dataset. Adopting a DL model then led to the use of augmentation strategies that produced a comprehensive dataset with consistent hyper parameters across all test cases. As a further step in the classification process, we used a Convolutional Neural Network model. Different enhancement methods have been used to raise visual quality. The proposed approach detected the DR with a highest experimental result of 97.83%, a top-2 accuracy of 99.31%, and a top-3 accuracy of 99.88% across all the 5 severity stages of the APTOS 2019 evaluation employing CLAHE and ESRGAN techniques for image enhancement. In addition, we employed APTOS 2019 to develop a set of evaluation metrics (precision, recall, and F1-score) to use in analyzing the efficacy of the suggested model. The proposed approach was also proven to be more efficient at DR location than both state-of-the-art technology and conventional DL.

Temporal deep learning framework for retinopathy prediction in patients with type 1 diabetes

The adoption of electronic health records in hospitals has ensured the availability of large datasets that can be used to predict medical complications. The trajectories of patients in real-world settings are highly variable, making longitudinal data modeling challenging. In recent years, significant progress has been made in the study of deep learning models applied to time series; however, the application of these models to irregular medical time series (IMTS) remains limited. To address this issue, we developed a generic deep-learning-based framework for modeling IMTS that facilitates the comparative studies of sequential neural networks (transformers and long short-term memory) and irregular time representation techniques. A validation study to predict retinopathy complications was conducted on 1207 patients with type 1 diabetes in a French database using their historical glycosylated hemoglobin measurements, without any data aggregation or imputation. The transformer-based model combined with the soft one-hot representation of time gaps achieved the highest score: an area under the receiver operating characteristic curve of 88.65%, specificity of 85.56%, sensitivity of 83.33% and an improvement of 11.7% over the same architecture without time information. This is the first attempt to predict retinopathy complications in patients with type 1 diabetes using deep learning and longitudinal data collected from patient visits. This study highlighted the significance of modeling time gaps between medical records to improve prediction performance and the utility of a generic framework for conducting extensive comparative studies.

Recent developments on computer aided systems for diagnosis of diabetic retinopathy: a review

Diabetes is a long-term condition in which the pancreas quits producing insulin or the body's insulin isn't utilised properly. One of the signs of diabetes is Diabetic Retinopathy. Diabetic retinopathy is the most prevalent type of diabetes, if remains unaddressed, diabetic retinopathy can affect all diabetics and become very serious, raising the chances of blindness. It is a chronic systemic condition that affects up to 80% of patients for more than ten years. Many researchers believe that if diabetes individuals are diagnosed early enough, they can be rescued from the condition in 90% of cases. Diabetes damages the capillaries, which are microscopic blood vessels in the retina. On images, blood vessel damage is usually noticeable. Therefore, in this study, several traditional, as well as deep learning-based approaches, are reviewed for the classification and detection of this particular diabetic-based eye disease known as diabetic retinopathy, and also the advantage of one approach over the other is also described. Along with the approaches, the dataset and the evaluation metrics useful for DR detection and classification are also discussed. The main finding of this study is to aware researchers about the different challenges occurs while detecting diabetic retinopathy using computer vision, deep learning techniques. Therefore, a purpose of this review paper is to sum up all the major aspects while detecting DR like lesion identification, classification and segmentation, security attacks on the deep learning models, proper categorization of datasets and evaluation metrics. As deep learning models are quite expensive and more prone to security attacks thus, in future it is advisable to develop a refined, reliable and robust model which overcomes all these aspects which are commonly found while designing deep learning models.

Multi-Model Domain Adaptation for Diabetic Retinopathy Classification

Diabetic retinopathy (DR) is one of the most threatening complications in diabetic patients, leading to permanent blindness without timely treatment. However, DR screening is not only a time-consuming task that requires experienced ophthalmologists but also easy to produce misdiagnosis. In recent years, deep learning techniques based on convolutional neural networks have attracted increasing research attention in medical image analysis, especially for DR diagnosis. However, dataset labeling is expensive work and it is necessary for existing deep-learning-based DR detection models. For this study, a novel domain adaptation method (multi-model domain adaptation) is developed for unsupervised DR classification in unlabeled retinal images. At the same time, it only exploits discriminative information from multiple source models without access to any data. In detail, we integrate a weight mechanism into the multi-model-based domain adaptation by measuring the importance of each source domain in a novel way, and a weighted pseudo-labeling strategy is attached to the source feature extractors for training the target DR classification model. Extensive experiments are performed on four source datasets (DDR, IDRiD, Messidor, and Messidor-2) to a target domain APTOS 2019, showing that MMDA produces competitive performance for present state-of-the-art methods for DR classification. As a novel DR detection approach, this article presents a new domain adaptation solution for medical image analysis when the source data is unavailable.

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.