No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Deep learning innovations in diagnosing diabetic retinopathy: The potential of transfer learning and the DiaCNN model
Abstract
Diabetic retinopathy (DR) is a significant cause of vision impairment, emphasizing the critical need for early detection and timely intervention to avert visual deterioration. Diagnosing DR is inherently complex, as it necessitates the meticulous examination of intricate retinal images by experienced specialists. This makes the early diagnosis of DR essential for effective treatment and prevention of eventual blindness. Traditional diagnostic methods, relying on human interpretation of these medical images, face challenges in terms of accuracy and efficiency. In the present research, we introduce a novel method that offers superior precision in DR diagnosis, compared to traditional methods, by employing advanced deep learning techniques. Central to this approach is the concept of transfer learning. This entails the utilization of pre-existing, well-established models, specifically InceptionResNetv2 and Inceptionv3, to extract features and fine-tune selected layers to cater to the unique requirements of this specific diagnostic task. Concurrently, we also present a newly devised model, DiaCNN, which is tailored for the classification of eye diseases. To prove the efficacy of the proposed methodology, we leveraged the Ocular Disease Intelligent Recognition (ODIR) dataset, which comprises eight different eye disease categories. The results are promising. The InceptionResNetv2 model, incorporating transfer learning, registered an impressive 97.5% accuracy in both the training and testing phases. Its counterpart, the Inceptionv3 model, achieved an even more commendable 99.7% accuracy during training, and 97.5% during testing. Remarkably, the DiaCNN model showcased unparalleled precision, achieving 100% accuracy in training and 98.3% in testing. These figures represent a significant leap in classification accuracy when juxtaposed with existing state-of-the-art diagnostic methods. Such advancements hold immense promise for the future, emphasizing the potential of our proposed technique to revolutionize the accuracy of DR and other eye disease diagnoses. By facilitating earlier detection and more timely interventions, this approach stands poised to significantly reduce the incidence of blindness associated with DR, thus heralding a new era of improved patient outcomes. Therefore, this work, through its novel approach and stellar results, not only pushes the boundaries of DR diagnostic accuracy but also promises a transformative impact in early detection and intervention, aiming to substantially diminish DR-induced blindness and champion enhanced patient care.
... Its prevalence is notably higher in regions where diabetes is more widespread. Approximately 60 million individuals are living with diabetes, with diabetic retinopathy impacting around 12% to 18% of diabetes patients in India [1]. Therefore, developing countries like India face a shortage of eye care facilities due to the substantial diabetic patient population. ...
According to the International Diabetes Federation (IDF), roughly 33% of individuals affected by diabetes exhibit diagnoses encompassing diverse severity of diabetic retinopathy. In the year 2020, approximately 463 million adults within the age bracket of 20 to 79 were documented as diabetes sufferers on a global scale. Projections suggest a rise to 700 million by 2045. Proposed automated diabetic retinopathy detection methods aim to reduce ophthalmologist workload. The study presents the Robust Fuzzy Local Information K-Means Clustering algorithm, an advanced iteration of the classical K-means clustering approach, integrating localized information parameters tailored to individual clusters. Comparative analysis is conducted between the performance of Robust Fuzzy Local Information K-Means Clustering and Modified Fuzzy C Means clustering, the latter of which incorporates a median adjustment parameter to augment Fuzzy C Means for diabetic retinopathy detection. The results are evaluated on three different datasets: IDRiD, Kaggle, and fundus images collected from Shiva Netralaya Center, India. Achieving a 94.4% accuracy rate and an average execution time of 17.11 seconds, the proposed algorithm aims to adeptly categorize a substantial volume of retinal images, thereby improving performance and meeting the crucial demand for prompt and precise diagnoses in diabetic retinopathy healthcare.
Automatic seizure detection and prediction using clinical Electroencephalograms (EEGs) are challenging tasks due to factors such as low Signal-to-Noise Ratios (SNRs), high variance in epileptic seizures among patients, and limited clinical data constraints. To overcome these challenges, this paper presents two approaches for EEG signal classification. One of these approaches depends on Machine Learning (ML) tools. The used features are different types of entropy, higher-order statistics, and sub-band energies in the Hilbert Marginal Spectrum (HMS) domain. The classification is performed using Support Vector Machine (SVM), Logistic Regression (LR), and K-Nearest Neighbor (KNN) classifiers. Both seizure detection and prediction scenarios are considered. The second approach depends on spectrograms of EEG signal segments and a Convolutional Neural Network (CNN)-based residual learning model. We use 10000 spectrogram images for each class. In this approach, it is possible to perform both seizure detection and prediction in addition to a 3-state classification scenario. Both approaches are evaluated on the Children's Hospital Boston and the Massachusetts Institute of Technology (CHB-MIT) dataset, which contains 24 EEG recordings for 6 males and 18 females. The results obtained for the HMS-based model showed an accuracy of 100%. The CNN-based model achieved accuracies of 97.66%, 95.59%, and 94.51% for Seizure (S) versus Pre-Seizure (PS), Non-Seizure (NS) versus S, and NS versus S versus PS classes, respectively. These results demonstrate that the proposed approaches can be effectively used for seizure detection and prediction. They outperform the state-of-the-art techniques for automatic seizure detection and prediction. Block diagram of proposed epileptic seizure detection method using HMS with different classifiers.
Acute lower respiratory infection is a leading cause of death in developing countries. Hence, progress has been made for early detection and treatment. There is still a need for improved diagnostic and therapeutic strategies, particularly in resource-limited settings. Chest X-ray and computed tomography (CT) have the potential to serve as effective screening tools for lower respiratory infections, but the use of artificial intelligence (AI) in these areas is limited. To address this gap, we present a computer-aided diagnostic system for chest X-ray and CT images of several common pulmonary diseases, including COVID-19, viral pneumonia, bacterial pneumonia, tuberculosis, lung opacity, and various types of carcinoma. The proposed system depends on super-resolution (SR) techniques to enhance image details. Deep learning (DL) techniques are used for both SR reconstruction and classification, with the InceptionResNetv2 model used as a feature extractor in conjunction with a multi-class support vector machine (MCSVM) classifier. In this paper, we compare the proposed model performance to those of other classification models, such as Resnet101 and Inceptionv3, and evaluate the effectiveness of using both softmax and MCSVM classifiers. The proposed system was tested on three publicly available datasets of CT and X-ray images and it achieved a classification accuracy of 98.028% using a combination of SR and InceptionResNetv2. Overall, our system has the potential to serve as a valuable screening tool for lower respiratory disorders and assist clinicians in interpreting chest X-ray and CT images. In resource-limited settings, it can also provide a valuable diagnostic support.
Brain tumor is an abnormal cell population that occurs in the brain. Currently, medical imaging techniques play a vital role in brain tumor diagnosis and classification. Brain tumor classification based on Magnetic Resonance Imaging (MRI) has become a promising research area in the field of medical imaging systems. In the brain image, the size of the tumor may vary from patient to patient along with the minute details of the tumor. It is a difficult task for radiologists to diagnose and classify tumors from numerous images. An efficient algorithm is proposed in this paper for tumor classification based on Deep Learning (DL) models. This paper presents three different Convolutional Neural Network (CNN) models for classification of brain tumors. These models are AlexNet, VGG16, and ResNet50. As brain images need to be stored for a along time for research and medical causes, image compression is an efficient tool for minimizing storage space, and also for allowing the deep analysis of brain images. This study depends on a lossy compression method, namely JPEG2000, for the storage of medical brain images. Classification is applied on the dataset with and without compression to estimate the effect of the compression method on the classification performance. Results of the classification models show that ResNet50 achieves a 99.97% accuracy, then VGG16 reaches a 98.83% accuracy, and finally, AlexNet gives a 92.92% accuracy without compression. The compression process is applied with four different compression ratios of 50, 25, 12.5, and 10%. The reduction in accuracy of classification with compression is small, as ResNet50 gives a 98.56% accuracy, and VGG16 gives a 92.92% accuracy, while AlexNet gives an 83.83% accuracy.
The brain tumor is an urgent malignancy caused by unregulated cell division. Tumors are classified using a biopsy, which is normally performed after the final brain surgery. Deep learning technology advancements have assisted the health professionals in medical imaging for the medical diagnosis of several symptoms. In this paper, transfer-learning-based models in addition to a Convolutional Neural Network (CNN) called BRAIN-TUMOR-net trained from scratch are introduced to classify brain magnetic resonance images into tumor or normal cases. A comparison between the pre-trained InceptionResNetv2, Inceptionv3, and ResNet50 models and the proposed BRAIN-TUMOR-net is introduced. The performance of the proposed model is tested on three publicly available Magnetic Resonance Imaging (MRI) datasets. The simulation results show that the BRAIN-TUMOR-net achieves the highest accuracy compared to other models. It achieves 100%, 97%, and 84.78% accuracy levels for three different MRI datasets. In addition, the k-fold cross-validation technique is used to allow robust classification. Moreover, three different unsupervised clustering techniques are utilized for segmentation.
Retinopathy is an eye disease caused by diabetes, and early detection and treatment can potentially reduce the risk of blindness in diabetic retinopathy sufferers. Using retinal Fundus images, diabetic retinopathy can be diagnosed, recognized, and treated. In the current state of the art, sensitivity and specificity are lacking. However, there are still a number of problems to be solved in state-of-the-art techniques like performance, accuracy, and being able to identify DR disease effectively with greater accuracy. In this paper, we have developed a new approach based on a combination of image processing and artificial intelligence that will meet the performance criteria for the detection of disease-causing diabetes retinopathy in Fundus images. Automatic detection of diabetic retinopathy has been proposed and has been carried out in several stages. The analysis was carried out in MATLAB using software-based simulation , and the results were then compared with those of expert ophthalmologists to verify their accuracy. Different types of diabetic retinopathy are represented in the experimental evaluation, including exudates, micro-aneurysms, and reti-nal hemorrhages. The detection accuracies shown by the experiments are greater than 98.80 percent.
This article focuses on automatic modulation classification (AMC) in wireless communication systems. A convolutional neural network (CNN) with three layers is introduced for the AMC process. Over degraded channels, it is assumed that the constellation diagrams of received signals do not show sharp points as in the case of pure signals. Instead, the points spread to constitute circle‐shaped objects. With more deterioration in channel conditions, these circle‐shaped objects begin to show overlapping. This behavior motivates us to use object detection, when dealing with the modulation classification task. The selection of the adopted transforms in this article is made from the object detection perspective. Different 2D transforms are considered on the constellation diagrams and compared for better classification performance. These transforms are the Radon transform (RT), the curvelet transform, and the phase congruency (PC). They are applied on the 2D constellation diagrams prior to the classification task with the CNN. The classification of the modulation format at different signal‐to‐noise ratios (SNRs) is considered in this article from the constellation diagrams, and the preprocessed constellation diagrams using RT, curvelet transform, and PC. Seven types of modulation formats are considered in this study to represent both spread and dense constellation diagram patterns, and the study extends from −10 to 10 dB. Analysis of the results indicating the most suitable preprocessor according to the constellation type and the SNR involved is provided.
Brain tumor is an acute cancerous disease that results from abnormal and uncontrollable cell division. Brain tumors are classified via biopsy, which is not normally done before the brain ultimate surgery. Recent advances and improvements in deep learning (DL) models helped the health industry in getting accurate diseases diagnosis. This article concentrates on the classification of magnetic resonance (MR) images. The objective is to differentiate between glioma tumors, meningioma tumors, pituitary tumors, and normal cases. Four deep convolutional neural networks are considered and compared in this article. These networks are inceptionresnetv2, inceptionv3, transfer learning, and BRAIN-TUMOR-net. A transfer-learning strategy is considered to enhance the performance of pre-trained models and save the time of training. The used dataset is the brain tumor magnetic resonance imaging dataset. It contains four classes including 826 MR images for glioma tumor, 822 MR images for meningioma tumor, 827 MR images for pituitary tumor, and 835 MR images for normal cases. Due to the limited number of images, we use the augmentation strategy to enlarge the size of the dataset. 75% of the data are considered for training and the other 25% are considered for testing. Segmentation of the classified results is performed. Simulation results prove that the DL model from scratch obtains the highest performance with the augmented data. In addition, a new practical implementation is presented for the proposed models.
This paper explores the issue of COVID-19 detection from X-ray images. X-ray images, in general, suffer from low quality and low resolution. That is why the detection of different diseases from X-ray images requires sophisticated algorithms. First of all, machine learning (ML) is adopted on the features extracted manually from the X-ray images. Twelve classifiers are compared for this task. Simulation results reveal the superiority of Gaussian process (GP) and random forest (RF) classifiers. To extend the feasibility of this study, we have modified the feature extraction strategy to give deep features. Four pre-trained models, namely ResNet50, ResNet101, Inception-v3 and InceptionResnet-v2 are adopted in this study. Simulation results prove that InceptionResnet-v2 and ResNet101 with GP classifier achieve the best performance. Moreover, transfer learning (TL) is also introduced in this paper to enhance the COVID-19 detection process. The selected classification hierarchy is also compared with a convolutional neural network (CNN) model built from scratch to prove its quality of classification. Simulation results prove that deep features and TL methods provide the best performance that reached 100% for accuracy.
Electroencephalography (EEG) is among the main tools used for analyzing and diagnosing epilepsy. The manual analysis of EEG must be conducted by highly trained clinicians or neuro-physiologists; a process that is considered to have a comparatively low inter-rater agreement. Furthermore, the new data interpretation consumes an excessive amount of time and resources. Hence, an automatic seizure detection and prediction system can improve the quality of patient care in terms of shortening the diagnosis period, reducing manual errors, and automatically detecting debilitating events. Moreover, for patient treatment, it is important to alert the patients of epilepsy seizures prior to seizure occurrence. Various distinguished studies presented good solutions for two-class seizure detection problems with binary classification scenarios. To deal with these challenges, this paper puts forward effective approaches for EEG signal classification for normal, pre-ictal, and ictal activities. Three models are presented for the classification task. Two of them are patient-specific, while the third one is patient non-specific, which makes it better for the general classification tasks. The two-class classification is implemented between normal and pre-ictal activities for seizure prediction and between normal and ictal activities for seizure detection. A more generalized three-class classification framework is considered to identify all EEG signal activities. The first model depends on a Convolutional Neural Network (CNN) with residual blocks. It contains thirteen layers with four residual learning blocks. It works on spectrograms of EEG signal segments. The second model depends on a CNN with three layers. It also works on spectrograms. On the other hand, the third model depends on Phase Space Reconstruction (PSR) to eliminate the limitations of the spectrograms used in the first models. A five-layer CNN is used with this strategy. The advantage of the PSR is the direct projection from the time domain, which keeps the main trend of different signal activities. The third model deals with all signal activities, and it was tested for all patients of the CHB-MIT dataset. It has a superior performance compared to the first models and the state-of-the-art models.
Brain tumor is an acute cancerous disease that results from abnormal and uncontrollable cell division. Brain tumors are classified via biopsy, which is not normally done before the brain ultimate surgery. Recent advances and improvements in deep learning technology helped the health industry in getting accurate disease diagnosis. In this paper, a Convolutional Neural Network (CNN) is adopted with image pre-processing to classify brain Magnetic Resonance (MR) images into four classes: glioma tumor, meningioma tumor, pituitary tumor and normal patients, is provided. We use a transfer learning model, a CNN-based model that is designed from scratch, a pre-trained inceptionresnetv2 model and a pre-trained inceptionv3 model. The performance of the four proposed models is tested using evaluation metrics including accuracy, sensitivity, specificity, precision, F1_score, Matthew’s correlation coefficient, error, kappa and false positive rate. The obtained results show that the two proposed models are very effective in achieving accuracies of 93.15% and 91.24% for the transfer learning model and BRAIN-TUMOR-net based on CNN, respectively. The inceptionresnetv2 model achieves an accuracy of 86.80% and the inceptionv3 model achieves an accuracy of 85.34%. Practical implementation of the proposed models is presented.
In recent years, much attention has been provided to developing the systems used for making medical diagnosis. That is because there are various developments made in the field of computing technology. The knowledge related to health and type of disease is essential for making a medical diagnosis by reliable and having an accurate detection. Regarding diabetic retinopathy (DR), it is a common cause for blindness. This disease can be detected in an early stage. It has various symptoms. The most distinctive symptoms are Micro-aneurysms (MAs), Hemorrhages (HAs) which are dark lesions and Hard Exudates (HEs), and Cotton Wool Spots (CWS) that are deemed as bright lesions. Through the present research, the researchers aimed at proposing an automated system for detecting exudates and cotton wool spots in an early stage and classifying them. Regarding the location of optic disc and the structure of blood vessels, they play a significant role in bright lesions for having the DR detected in an early stage. This paper presents algorithms and techniques for processing the retinal image, blood vessel segmentation and extraction, optic disc localization and removal, feature extraction and finally classification for different bright lesions using SVM and Naïve-Bayes classifiers. For testing retinal images, the researchers used image-Ret database which is a publicly available database. This database includes two sub-databases (i.e., DIARETDB0, and DIARETDB1). Finally, we compare the performance of recently published works and our proposed work.
This article is mainly concerned with COVID-19 diagnosis from X-ray images. The number of cases infected with COVID-19 is increasing daily, and there is a limitation in the number of test kits needed in hospitals. Therefore, there is an imperative need to implement an efficient automatic diagnosis system to alleviate COVID-19 spreading among people. This article presents a discussion of the utilization of convolutional neural network (CNN) models with different learning strategies for automatic COVID-19 diagnosis. First, we consider the CNN-based transfer learning approach for automatic diagnosis of COVID-19 from X-ray images with different training and testing ratios. Different pre-trained deep learning models in addition to a transfer
learning model are considered and compared for the task of COVID-19 detection from X-ray images. Confusion matrices of these studied models are presented and analyzed. Considering the performance results obtained, ResNet models (ResNet18, ResNet50, and ResNet101) provide the highest classification accuracy on the two
considered datasets with different training and testing ratios, namely 80/20, 70/30, 60/40, and 50/50. The accuracies obtained using the first dataset with 70/30 training and testing ratio are 97.67%, 98.81%, and 100% for ResNet18, ResNet50, and ResNet101, respectively. For the second dataset, the reported accuracies are 99%, 99.12%, and 99.29% for ResNet18, ResNet50, and ResNet101, respectively. The second approach is the training of a proposed CNN model from scratch. The results confirm that training of the CNN from scratch can lead to the identification of the signs of COVID-19 disease.
Diabetic retinopathy is one of the major microvascular complications of diabetes mellitus. The most common causes of vision loss in diabetic retinopathy are diabetic macular edema and proliferative diabetic retinopathy. Recent developments in ocular imaging have played a significant role in early diagnosis and management of these complications. Color fundus photography is an imaging modality, which is helpful for screening patients with diabetic eye disease and monitoring its progression as well as response to treatment. Fundus fluorescein angiography (FFA) is a dye-based invasive test to detect subtle neovascularization, look for areas of capillary non-perfusion, diagnose macular ischemia, and differentiate between focal and diffuse capillary bed leak in cases of macular edema. Recent advances in retinal imaging like the introduction of spectral-domain and swept source-based optical coherence tomography (OCT), fundus autofluorescence (FAF), OCT angiography, and ultrawide field imaging and FFA have helped clinicians in the detection of certain biomarkers that can identify disease at an early stage and predict response to treatment in diabetic macular edema. This article will summarize the role of different imaging biomarkers in characterizing diabetic retinopathy and their potential contribution in its management.
Deep learning of fundus photograph has emerged as a practical and cost-effective technique for automatic screening and diagnosis of severer diabetic retinopathy (DR). The entropy image of luminance of fundus photograph has been demonstrated to increase the detection performance for referable DR using a convolutional neural network- (CNN-) based system. In this paper, the entropy image computed by using the green component of fundus photograph is proposed. In addition, image enhancement by unsharp masking (UM) is utilized for preprocessing before calculating the entropy images. The bichannel CNN incorporating the features of both the entropy images of the gray level and the green component preprocessed by UM is also proposed to improve the detection performance of referable DR by deep learning.
This proposal tempts to develop automated DR detection by analyzing the retinal abnormalities like hard exudates, haemorrhages, Microaneurysm, and soft exudates. The main processing phases of the developed DR detection model is Pre-processing, Optic Disk removal, Blood vessel removal, Segmentation of abnormalities, Feature extraction, Optimal feature selection, and Classification. At first, the pre-processing of the input retinal image is done by Contrast Limited Adaptive Histogram Equalization. The next phase performs the optic disc removal, which is carried out by open-close watershed transformation. Further, the Grey Level thresholding is done for segmenting the blood vessels and its removal. Once the optic disk and blood vessels are removed, segmentation of abnormalities is done by Top hat transformation and Gabor filtering. Further, the feature extraction phase is started, which tends to extract four sets of features like Local Binary Pattern, Texture Energy Measurement, Shanon’s and Kapur’s entropy. Since the length of the feature vector seems to be long, the feature selection process is done, which selects the unique features with less correlation. Moreover, the Deep Belief Network (DBN)-based classification algorithm performs the categorization of images into four classes normal, earlier, moderate, or severe stages. The optimal feature selection is done by the improved meta-heuristic algorithm called Modified Gear and Steering-based Rider Optimization Algorithm (MGS-ROA), and the same algorithm updates the weight in DBN. Finally, the effectual performance and comparative analysis prove the stable and reliable performance of the proposed model over existing models. The performance of the proposed model is compared with the existing classifiers, such as, NN, KNN, SVM, DBN and the conventional Heuristic-Based DBNs, such as PSO-DBN, GWO-DBN, WOA-DBN, and ROA-DBN for the evaluation metrics, accuracy, sensitivity, specificity, precision, FPR, FNR, NPV, FDR, F1 score, and MC. From the results, it is exposed that the accuracy of the proposed MGS-ROA-DBN is 30.1% higher than NN, 32.2% higher than KNN, and 17.1% higher than SVM and DBN. Similarly, the accuracy of the developed MGS-ROA-DBN is 13.8% superior to PSO, 5.1% superior to GWO, 10.8% superior to WOA, and 2.5% superior to ROA.
Purpose
To review the current therapeutic options for the management of diabetic retinopathy (DR) and diabetic macular edema (DME) and examine the evidence for integration of laser and pharmacotherapy.
Methods
A review of the PubMed database was performed using the search terms diabetic retinopathy, diabetic macular edema, neovascularization, laser photocoagulation, intravitreal injection, vascular endothelial growth factor (VEGF), vitrectomy, pars plana vitreous surgery, antiangiogenic therapy. With additional cross-referencing, this yielded 835 publications of which 301 were selected based on content and relevance.
Results
Many recent studies have evaluated the pharmacological, laser and surgical therapeutic strategies for the treatment and prevention of DR and DME. Several newer diagnostic systems such as optical coherence tomography (OCT), microperimetry, and multifocal electroretinography (mfERG) are also assisting in further refinements in the staging and classification of DR and DME. Pharmacological therapies for both DR and DME include both systemic and ocular agents. Systemic agents that promote intensive glycemic control, control of dyslipidemia and antagonists of the renin-angiotensin system demonstrate beneficial effects for both DR and DME. Ocular therapies include anti-VEGF agents, corticosteroids and nonsteroidal anti-inflammatory drugs. Laser therapy, both as panretinal and focal or grid applications continue to be employed in management of DR and DME. Refinements in laser devices have yielded more tissue-sparing (subthreshold) modes in which many of the benefits of conventional continuous wave (CW) lasers can be obtained without the adverse side effects. Recent attempts to lessen the burden of anti-VEGF injections by integrating laser therapy have met with mixed results. Increasingly, vitreoretinal surgical techniques are employed for less advanced stages of DR and DME. The development and use of smaller gauge instrumentation and advanced anesthesia agents have been associated with a trend toward earlier surgical intervention for diabetic retinopathy. Several novel drug delivery strategies are currently being examined with the goal of decreasing the therapeutic burden of monthly intravitreal injections. These fall into one of the five categories: non-biodegradable polymeric drug delivery systems, biodegradable polymeric drug delivery systems, nanoparticle-based drug delivery systems, ocular injection devices and with sustained release refillable devices. At present, there remains no one single strategy for the management of the particular stages of DR and DME as there are many options that have not been rigorously tested through large, randomized, controlled clinical trials.
Conclusion
Pharmacotherapy, both ocular and systemic, will be the primary mode of intervention in the management of DR and DME in many cases when cost and treatment burden are less constrained. Conventional laser therapy has become a secondary intervention in these instances, but remains a first-line option when cost and treatment burden are more constrained. Results with subthreshold laser appear promising but will require more rigorous study to establish its role as adjunctive therapy. Evidence to support an optimal integration of the various treatment options is lacking. Central to the widespread adoption of any therapeutic regimen for DR and DME is substantiation of safety, efficacy, and cost-effectiveness by a body of sound clinical trials.
Diabetes mellitus have an eye disease called diabetic retinopathy. The early discovery of the disease is a great achievement in management of diabetic retinopathy. We use Fundus images are used for identification of the nature of an illness or other problem through examination of the symptoms to check for any abnormalities or any change in the retina. In this paper, Convolutional Neural Networks (CNN) is performed to classify the retinal fundus images to normal, background and pre-proliferative retinopathy. The proposed model consists of 5 convolutional layers followed by 5 max pooling layers. Finally, a global average pooling is used. In this work we achieve accuracy reached 95.23%.
Purpose:
To achieve automatic diabetic retinopathy (DR) detection in retinal fundus photographs through the use of a deep transfer learning approach using the Inception-v3 network.
Methods:
A total of 19,233 eye fundus color numerical images were retrospectively obtained from 5278 adult patients presenting for DR screening. The 8816 images passed image-quality review and were graded as no apparent DR (1374 images), mild nonproliferative DR (NPDR) (2152 images), moderate NPDR (2370 images), severe NPDR (1984 images), and proliferative DR (PDR) (936 images) by eight retinal experts according to the International Clinical Diabetic Retinopathy severity scale. After image preprocessing, 7935 DR images were selected from the above categories as a training dataset, while the rest of the images were used as validation dataset. We introduced a 10-fold cross-validation strategy to assess and optimize our model. We also selected the publicly independent Messidor-2 dataset to test the performance of our model. For discrimination between no referral (no apparent DR and mild NPDR) and referral (moderate NPDR, severe NPDR, and PDR), we also computed prediction accuracy, sensitivity, specificity, area under the receiver operating characteristic curve (AUC), and κ value.
Results:
The proposed approach achieved a high classification accuracy of 93.49% (95% confidence interval [CI], 93.13%-93.85%), with a 96.93% sensitivity (95% CI, 96.35%-97.51%) and a 93.45% specificity (95% CI, 93.12%-93.79%), while the AUC was up to 0.9905 (95% CI, 0.9887-0.9923) on the independent test dataset. The κ value of our best model was 0.919, while the three experts had κ values of 0.906, 0.931, and 0.914, independently.
Conclusions:
This approach could automatically detect DR with excellent sensitivity, accuracy, and specificity and could aid in making a referral recommendation for further evaluation and treatment with high reliability.
Translational relevance:
This approach has great value in early DR screening using retinal fundus photographs.
Diabetic Retinopathy (DR) is an ophthalmic disease that damages retinal blood vessels. DR causes impaired vision and may even lead to blindness if it is not diagnosed in early stages. DR has five stages or classes, namely normal, mild, moderate, severe and PDR (Proliferative Diabetic Retinopathy). Normally, highly trained experts examine the colored fundus images to diagnose this fatal disease. This manual diagnosis of this condition (by clinicians) is tedious and error-prone. Therefore, various computer vision-based techniques have been proposed to automatically detect DR and its different stages from retina images. However, these methods are unable to encode the underlying complicated features and can only classify DR’s different stages with very low accuracy particularly, for the early stages. In this research, we used the publicly available Kaggle dataset of retina images to train an ensemble of five deep Convolution Neural Network (CNN) models (Resnet50, Inceptionv3, Xception, Dense121, Dense169) to encode the rich features and improve the classification for different stages of DR. The experimental results show that the proposed model detects all the stages of DR unlike the current methods and performs better compared to state-of-the-art methods on the same Kaggle dataset.
Diabetic retinopathy (DR) is an important causes of blindness worldwide. However, DR is hard to detected in early stages and the diagnostic procedure can be time-consuming even for experienced experts. Therefore, a computer-aided diagnosis method based on deep learning algorithms is proposed to automatedly diagnose the referable deiabetic retinopathy (RDR) by classifying color retinal fundus photographs into two grades. In this work, A novel convolutional neural network model with Siamese-like architecture is trained with transfer learning technique. Different from previous works, the proposed model accepts binocular fundus images as inputs and learns their correlation to help making prediction. In the case with a training set of only 28104 images and a test set of 7024 images, an area under the receiver operating curve (AUC) of 0.951 is obtained by the proposed binocular model, which is 0.011 higher than that obtained by existing monocular model. To further verify the effectiveness of the binocular design, a binocular model for five-class DR detection is also trained and evaluated on a 10% validation set. The result shows that it achieves a kappa score of 0.829 which is higher than that of existing non-emsemble model.
Diabetic retinopathy (DR) is a complication of long-standing diabetes, which is hard to detect in its early stage because it only shows a few symptoms. Nowadays, the diagnosis of DR usually requires taking digital fundus images, as well as images using optical coherence tomography (OCT). Since OCT equipment is very expensive, it will benefit both the patients and the ophthalmologists if an accurate diagnosis can be made, based solely on reading digital fundus images. In the paper, we present a novel algorithm based on deep convolutional neural network (DCNN). Unlike the traditional DCNN approach, we replace the commonly used max-pooling layers with fractional max-pooling. Two of these DCNNs with a different number of layers are trained to derive more discriminative features for classification. After combining features from metadata of the image and DCNNs, we train a support vector machine (SVM) classifier to learn the underlying boundary of distributions of each class. For the experiments, we used the publicly available DR detection database provided by Kaggle. We used 34,124 training images and 1,000 validation images to build our model and tested with 53,572 testing images. The proposed DR classifier classifies the stages of DR into five categories, labeled with an integer ranging between zero and four. The experimental results show that the proposed method can achieve a recognition rate up to 86.17%, which is higher than previously reported in the literature. In addition to designing a machine learning algorithm, we also develop an app called “Deep Retina.” Equipped with a handheld ophthalmoscope, the average person can take fundus images by themselves and obtain an immediate result, calculated by our algorithm. It is beneficial for home care, remote medical care, and self-examination.
Diabetic retinopathy (DR) is a common eye disease and a significant cause of blindness in diabetic patients. Regular screening with fundus photography and timely intervention is the most effective way to manage the disease. The large population of diabetic patients and their massive screening requirements have generated interest in computer-aided and fully automatic diagnosis of DR. Deep neural networks, on the other hand, have brought many breakthroughs in various tasks in the recent years. To automate the diagnosis of DR and provide the appropriate suggestions to DR patients, we have built a dataset of DR fundus images that have been labeled by the proper treatment method that is required. Using this dataset, we trained deep convolutional neural network models to grade the severities of DR fundus images. We were able to achieve an accuracy of 88.72% for a four degree classification task in the experiments. We deployed our models on a cloud computing platform and provided pilot DR diagnostic services for several hospitals, in the clinical evaluation the system achieved a consistency rate of 91.8% with ophthalmologists, demonstrating the effectiveness of our work.
Diabetic retinopathy (i.e., DR), is an eye disorder caused by diabetes, diabetic retinopathy detection is an important task in retinal fundus images due the early detection and treatment can potentially reduce the risk of blindness. Retinal fundus images play an important role in diabetic retinopathy through disease diagnosis, disease recognition (i.e., by ophthalmologists), and treatment. The current state-of-the-art techniques are not satisfied with sensitivity and specificity. In fact, there are still other issues to be resolved in state-of-the-art techniques such as performances, accuracy, and easily identify the DR disease effectively. Therefore, this paper proposes an effective image processing method for detection of diabetic retinopathy diseases from retinal fundus images that will satisfy the performance metrics (i.e., sensitivity, specificity, accuracy). The proposed automatic screening system for diabetic retinopathy was conducted in several steps: Pre-processing, optic disc detection and removal, blood vessel segmentation and removal, elimination of fovea, feature extraction (i.e., Microaneurysm, retinal hemorrhage, and exudates), feature selection and classification. Finally, a software-based simulation using MATLAB was performed using DIARETDB1 dataset and the obtained results are validated by comparing with expert ophthalmologists. The results of the conducted experiments showed an efficient and effective in sensitivity, specificity and accuracy.
Diabetic retinopathy (i.e., DR), is an eye disorder caused by diabetes, diabetic retinopathy detection is an important task in retinal fundus images due the early detection and treatment can potentially reduce the risk of blindness. Retinal fundus images play an important
role in diabetic retinopathy through disease diagnosis, disease recognition (i.e., by ophthalmologists), and treatment. The current state-of-the-art techniques are not satisfied with sensitivity and specificity. In fact, there are still other issues to be resolved in state-of-the-art techniques such as performances, accuracy, and easily identify the DR disease effectively. Therefore, this paper proposes an effective image processing method for detection of diabetic retinopathy diseases from retinal fundus images that will satisfy the performance metrics (i.e., sensitivity, specificity, accuracy). The proposed automatic screening system for diabetic retinopathy was conducted in several steps: Pre-processing, optic disc detection and removal, blood vessel segmentation and removal, elimination of fovea, feature extraction (i.e., Micro-aneurysm, retinal hemorrhage, and exudates), feature selection and classification. Finally, a software-based simulation using MATLAB was performed using DIARETDB1 dataset and the obtained results are validated by comparing with expert ophthalmologists. The results of the conducted experiments showed an efficient and effective in sensitivity, specificity and accuracy.
p>Diabetic retinopathy is a major cause of blindness in the United States. With rise of the epidemic of obesity and diabetes in the USA and around the globe, serious and common diabetic complications are
evolving as a major public health problem, particularly among minority populations. These populations
are disproportionately affected by diabetes and 2-3 times more likely to develop visually signifi cant
complications. In this highly illustrated review article, we discuss the diabetic epidemic, highlighting
the biology and the pathophysiologic mechanisms of this disorder on the anatomy of the eye. We also
discuss the risk factors and the implications for minority populations. For the health care providers, we
provide cutting edge information and imminently relevant information to help evaluate, manage, and know
when to refer their patients to a specialist in ophthalmology to quell the tide of the epidemic.
Pubmed link: https://www.ncbi.nlm.nih.gov/pubmed/29756128 </p
Diabetes or more precisely Diabetes Mellitus (DM) is a metabolic disorder happens because of high blood sugar level in the body. Over the time, diabetes creates eye deficiency also called as Diabetic Retinopathy (DR) causes major loss of vision. The symptoms can originate in the retinal area are augmented blood vessels, fluid drip, exudates, hemorrhages, and micro aneurysms. In modern medical science, images are the indispensable tool for precise diagnosis of patients. In the meantime evaluation of contemporary medical imageries remains complex. In recent times computer vision with Deep Neural Networks can train a model perfectly and level of accuracy also will be higher than other neural network models. In this study fundus images containing diabetic retinopathy has been taken into consideration. The idea behind this paper is to propose an automated knowledge model to identify the key antecedents of DR. Proposed Model have been trained with three types, back propagation NN, Deep Neural Network (DNN) and Convolutional Neural Network (CNN) after testing models with CPU trained Neural network gives lowest accuracy because of one hidden layers whereas the deep learning models are out performing NN. The Deep Learning models are capable of quantifying the features as blood vessels, fluid drip, exudates, hemorrhages and micro aneurysms into different classes. Model will calculate the weights which gives severity level of the patient's eye. The foremost challenge of this study is the accurate verdict of each feature class thresholds. To identify the target class thresholds weighted Fuzzy C-means algorithm has been used. The model will be helpful to identify the proper class of severity of diabetic retinopathy images.
In the last two decades Computer Aided Diagnostics (CAD) systems were developed to help radiologists analyze screening mammograms. The benefits of current CAD technologies appear to be contradictory and they should be improved to be ultimately considered useful. Since 2012 deep convolutional neural networks (CNN) have been a tremendous success in image recognition, reaching human performance. These methods have greatly surpassed the traditional approaches, which are similar to currently used CAD solutions. Deep CNN-s have the potential to revolutionize medical image analysis. We propose a CAD system based on one of the most successful object detection frameworks, Faster R-CNN. The system detects and classifies malignant or benign lesions on a mammogram without any human intervention. Our approach described here has achieved the 2nd place in the Digital Mammography DREAM Challenge with $ AUC = 0.85 $. The proposed method also sets the state of the art classification performance on the public INbreast database, $ AUC = 0.95$. When used as a detector, the system reaches high sensitivity with very few false positive marks per image on the INbreast dataset.
Diabetes is one of the foremost causes for the increase in mortality among children and adults in recent
years. Classification systems are being used by doctors to analyse and diagnose the medical data. Radial
basis function neural networks are more attractive for classification of diseases, especially in diabetes
classification, because of itrs non iterative nature. Radial basis function neural networks are four layer
feed forward neural network with input layer, pattern layer, summation layer and the decision layer respectively.
The size of the pattern layer increases on par with training data set size. Though various
attempts have made to solve this issue by clustering input data using different clustering algorithms like
k-means, k-medoids, and SOFM etc. However main difficulty of determining the optimal number of neurons
in the pattern layer remain unsolved. In this paper, we present a new model based on cluster validity
index with radial basis neural network for classification of diabetic patients data. We employ cluster validity
index in class by class fashion for determining the optimal number of neurons in pattern layer. A
new convex fitness function has also been designed for bat inspired optimization algorithm to identify the
weights between summation layer and pattern layer. The proposed model for radial basis function neural
network is tested on Pima Indians Diabetes data set and synthetic data sets. Experimental results proved
that our approach performs better in terms of accuracy, sensitivity, specificity, classification time, training
time, network complexity and computational time compared to conventional radial basis function neural
network. It is also proved that proposed model performs better compared to familiar classifiers namely
probabilistic neural network, feed forward neural network, cascade forward network, time delay network,
artificial immuine system and GINI classifier.
Purpose:
Diabetes is a major public health problem affecting 415 million people worldwide. With the increasing prevalence of diabetes, diabetic retinopathy (DR) is emerging as the leading cause of avoidable blindness worldwide.
Methods:
We reviewed previous and recent literature to provide an overview of emerging trends on the burden, epidemiology, risk factors, and prevention of DR.
Results:
First, there is clear evidence of a global increase in the prevalence of diabetes. Second, there is a decline in the incidence of blindness due to DR, particularly in developed countries. Third, diabetic macular edema (DME) rather than proliferative diabetic retinopathy (PDR) is the increasingly common cause of visual impairment. Fourth, DR awareness remains patchy and low in most populations. Fifth, hyperglycemia remains the most consistent risk factor for DR in type 1 diabetes across different studies and populations. Sixth, in contrast, blood pressure is an important risk factor for DR in type 2 diabetes. Seventh, the relationship between dyslipidemia and DR remains unclear, with inconsistent results from different studies and trials. Eighth, the utility of predictive models incorporating multiple risk factors for assessing DR risk requires evaluation. Ninth, photographic screening of DR using tele-ophthalmology platforms is increasingly recognized as being feasible and cost-effective. Finally, DR prevention in low-resource settings cannot follow models developed in high-resource countries and requires different strategies.
Conclusions:
The ten trends we observed in the current review may guide planning of public healthcare strategies for the management of DR and prevention of blindness.
Remarkable progress has been made in image recognition, primarily due to the availability of large-scale annotated datasets and the revival of deep CNN. CNNs enable learning data-driven, highly representative, layered hierarchical image features from sufficient training data. However, obtaining datasets as comprehensively annotated as ImageNet in the medical imaging domain remains a challenge. There are currently three major techniques that successfully employ CNNs to medical image classification: training the CNN from scratch, using off-the-shelf pre-trained CNN features, and conducting unsupervised CNN pre-training with supervised fine-tuning. Another effective method is transfer learning, i.e., fine-tuning CNN models pre-trained from natural image dataset to medical image tasks. In this paper, we exploit three important, but previously understudied factors of employing deep convolutional neural networks to computer-aided detection problems. We first explore and evaluate different CNN architectures. The studied models contain 5 thousand to 160 million parameters, and vary in numbers of layers. We then evaluate the influence of dataset scale and spatial image context on performance. Finally, we examine when and why transfer learning from pre-trained ImageNet (via fine-tuning) can be useful. We study two specific computer-aided detection (CADe) problems, namely thoraco-abdominal lymph node (LN) detection and interstitial lung disease (ILD) classification. We achieve the state-of-the-art performance on the mediastinal LN detection, with 85% sensitivity at 3 false positive per patient, and report the first five-fold cross-validation classification results on predicting axial CT slices with ILD categories. Our extensive empirical evaluation, CNN model analysis and valuable insights can be extended to the design of high performance CAD systems for other medical imaging tasks.
It has been past more than 15 years since the F-measure was first introduced to evaluation tasks of information extraction technology at the Fourth Message Understanding Conference (MUC-4) in 1992. Recently, sometimes I see some confusion with the definition of the F-measure, which seems to be triggered by lack of background knowledge about how the F-measure was derived. Since I was not involved in the process of the introduction or device of the F-measure, I might not be the best person to explain this but I hope this note would be a little help for those who are wondering what the F-measure really is. This introduction is devoted to provide brief but sufficient information on the F-measure.
We propose a deep convolutional neural network architecture codenamed
"Inception", which was responsible for setting the new state of the art for
classification and detection in the ImageNet Large-Scale Visual Recognition
Challenge 2014 (ILSVRC 2014). The main hallmark of this architecture is the
improved utilization of the computing resources inside the network. This was
achieved by a carefully crafted design that allows for increasing the depth and
width of the network while keeping the computational budget constant. To
optimize quality, the architectural decisions were based on the Hebbian
principle and the intuition of multi-scale processing. One particular
incarnation used in our submission for ILSVRC 2014 is called GoogLeNet, a 22
layers deep network, the quality of which is assessed in the context of
classification and detection.
This paper presents the development and evaluation of a fraud detection model for identifying fraudulent cases and anomalous instances in a business system. The authors faced a dilemma between choosing a univariate model (Isolation Forest) or a multivariate model (Random Cut Forest) to identify fraudulent cases and ultimately decided on the multivariate model due to its ability to detect all types of anomalies and its safety for the system. The model was evaluated using a limited number of confirmed fraudulent cases and a cutoff score of 1.4 and was able to identify instances similar to the known cases of fraud. In addition to identifying confirmed cases of fraud, the model also identified anomalies that required further investigation, which was essential in building a more effective supervised model and improving the system’s accuracy in identifying future instances of fraud. The findings from the investigations were also used to refine the engine rules, strengthening the overall security and stability of the system. Although the abstract does not mention the specific context of a food delivery business or Amazon, the paper describes a fraud detection model that could be applied in various business systems. The paper contributes to the current state of knowledge by presenting a reliable and effective fraud detection model that improves the overall security and stability of a business system.
Deep learning classifiers provide the most accurate means of automatically diagnosing diabetic retinopathy (DR) based on optical coherence tomography (OCT) and its angiography (OCTA). The power of these models is attributable in part to the inclusion of hidden layers that provide the complexity required to achieve a desired task. However, hidden layers also render algorithm outputs difficult to interpret. Here we introduce a novel biomarker activation map (BAM) framework based on generative adversarial learning that allows clinicians to verify and understand classifiers decision-making. A data set including 456 macular scans were graded as non-referable or referable DR based on current clinical standards. A DR classifier that was used to evaluate our BAM was first trained based on this data set. The BAM generation framework was designed by combing two U-shaped generators to provide meaningful interpretability to this classifier. The main generator was trained to take referable scans as input and produce an output that would be classified by the classifier as non-referable. The BAM is then constructed as the difference image between the output and input of the main generator. To ensure that the BAM only highlights classifier-utilized biomarkers an assistant generator was trained to do the opposite, producing scans that would be classified as referable by the classifier from non-referable scans. The generated BAMs highlighted known pathologic features including nonperfusion area and retinal fluid. A fully interpretable classifier based on these highlights could help clinicians better utilize and verify automated DR diagnosis.
Though deep learning has shown successful performance in classifying the label and severity stage of certain diseases, most of them give few explanations on how to make predictions. Inspired by Koch's Postulates, the foundation in evidence-based medicine (EBM) to identify the pathogen, we propose to exploit the interpretability of deep learning application in medical diagnosis. By isolating neuron activation patterns from a diabetic retinopathy (DR) detector and visualizing them, we can determine the symptoms that the DR detector identifies as evidence to make prediction. To be specific, we first define novel pathological descriptors using activated neurons of the DR detector to encode both spatial and appearance information of lesions. Then, to visualize the symptom encoded in the descriptor, we propose
Patho-GAN
, a new network to synthesize medically plausible retinal images. By manipulating these descriptors, we could even arbitrarily control the position, quantity, and categories of generated lesions. We also show that our synthesized images carry the symptoms directly related to diabetic retinopathy diagnosis. Our generated images are both qualitatively and quantitatively superior to the ones by previous methods. Besides, compared to existing methods that take hours to generate an image, our second level speed endows the potential to be an effective solution for data augmentation.
Diabetic Retinopathy (DR) is a complication of long-standing, unchecked diabetes, and one of the leading causes of blindness in the world. This paper focuses on improved and robust methods to extract some of the features of DR, viz., Blood Vessels and Exudates. Blood vessels are segmented using multiple morphological and thresholding operations. For the segmentation of exudates, k-means clustering and contour detection on the original images are used. Extensive noise reduction is performed to remove false positives from the vessel segmentation algorithm’s results. The localization of optic disc using k-means clustering and template matching is also performed. Lastly, this paper presents a Deep Convolutional Neural Network (DCNN) model with 14 Convolutional Layers and 2 Fully Connected Layers, for the automatic, binary diagnosis of DR. The vessel segmentation, optic disc localization and DCNN achieve accuracies of 95.93%, 98.77%, and 75.73%, respectively.
In recent years, Artificial Intelligence (AI) has proven its relevance for medical decision support. However, the “black-box” nature of successful AI algorithms still holds back their wide-spread deployment. In this paper, we describe an eXplanatory Artificial Intelligence (XAI) that reaches the same level of performance as black-box AI, for the task of classifying Diabetic Retinopathy (DR) severity using Color Fundus Photography (CFP). This algorithm, called ExplAIn, learns to segment and categorize lesions in images; the final image-level classification directly derives from these multivariate lesion segmentations. The novelty of this explanatory framework is that it is trained from end to end, with image supervision only, just like black-box AI algorithms: the concepts of lesions and lesion categories emerge by themselves. For improved lesion localization, foreground/background separation is trained through self-supervision, in such a way that occluding foreground pixels transforms the input image into a healthy-looking image. The advantage of such an architecture is that automatic diagnoses can be explained simply by an image and/or a few sentences. ExplAIn is evaluated at the image level and at the pixel level on various CFP image datasets. We expect this new framework, which jointly offers high classification performance and explainability, to facilitate AI deployment.
Diabetic retinopathy is one of the major microvascular complications of diabetes mellitus. The most common causes of vision loss in diabetic retinopathy are diabetic macular edema and proliferative diabetic retinopathy. Recent developments in ocular imaging have played a significant role in early diagnosis and management of these complications. Color fundus photography is an imaging modality, which is helpful for screening patients with diabetic eye disease and monitoring its progression as well as response to treatment. Fundus fluorescein angiography (FFA) is a dye-based invasive test to detect subtle neovascularization, look for areas of capillary non-perfusion, diagnose macular ischemia, and differentiate between focal and diffuse capillary bed leak in cases of macular edema. Recent advances in retinal imaging like the introduction of spectral-domain and swept source-based optical coherence tomography (OCT), fundus autofluorescence (FAF), OCT angiography, and ultrawide field imaging and FFA have helped clinicians in the detection of certain biomarkers that can identify disease at an early stage and predict response to treatment in diabetic macular edema. This article will summarize the role of different imaging biomarkers in characterizing diabetic retinopathy and their potential contribution in its management.
The early detection of diabetic retinopathy is crucial for preventing blindness. However, it is time-consuming to analyze fundus images manually, especially considering the increasing amount of medical images. In this paper, we propose an automatic diabetic retinopathy screening method using color fundus images. Our approach consists of three main components: edge-guided candidate microaneurysms detection, candidates classification using mixed features, and diabetic retinopathy prediction using fused features of image level and lesion level. We divide a screening task into two sub-classification tasks: 1) verifying candidate microaneurysms by a naive Bayes classifier; 2) predicting diabetic retinopathy using a support vector machine classifier. Our approach can effectively alleviate the imbalanced class distribution problem. We evaluate our method on two public databases: Lariboisìere and Messidor, resulting in an area under the curve of 0.908 on Lariboisìere and 0.832 on Messidor. These scores demonstrate the advantages of our approach over the existing methods.
In this study, a deep-transfer learning approach is proposed for the automated diagnosis of diabetes mellitus (DM), using heart rate (HR) signals obtained from electrocardiogram (ECG) data. Recent progress in deep learning has contributed significantly to improvement in the quality of healthcare. In order for deep learning models to perform well, large datasets are required for training. However, a difficulty in the biomedical field is the lack of clinical data with expert annotation. A recent, commonly implemented technique to train deep learning models using small datasets is to transfer the weighting, developed from a large dataset, to the current model. This deep learning transfer strategy is generally employed for two-dimensional signals. Herein, the weighting of models pre-trained using two-dimensional large image data was applied to one-dimensional HR signals. The one-dimensional HR signals were then converted into frequency spectrum images, which were utilized for application to well-known pre-trained models, specifically: AlexNet, VggNet, ResNet and DenseNet. The DenseNet pre-trained model yielded the highest classification average accuracy of 97.62%, and sensitivity of 100%, to detect DM subjects via HR signal recordings. In the future, we intend to further test this developed model by utilizing additional data along with cloud-based storage, in order to diagnose DM via heart signal analysis.
Aims
Diabetic retinopathy (DR) is an important microvascular complication of diabetes mellitus (DM) and a leading cause of visual impairment and blindness among people of working age. Physical activity (PA) or exercise is critical and beneficial for DM patients, whereas studies evaluating the relationship between PA and DR have yielded inconsistent and inconclusive results. The American Diabetes Association’s “Standards of Medical Care in Diabetes” has also pointed out the indeterminate roles of PA in DR prevention. The purpose of this systematic review and meta-analysis was to explore the association between PA and DR risk.
Methods
Medline (accessed by PubMed), EmBase, and Cochrane Library were systematically searched for studies up to June 2018, and the reference lists of the published articles were searched manually. The association between PA and DR risk was assessed using random-effect meta-analysis.
Results
Twenty-two studies were included in this meta-analysis. PA was found to have a protective association with DR [risk ratio (RR) = 0.94, 95% confidence interval (95% CI) 0.90–0.98, p = 0.005] in diabetic patients, and the impact was more pronounced on vision-threatening DR (RR = 0.89, 95% CI 0.80–0.98, p = 0.02). Sedentary behavior could increase the risk of DR (RR = 1.18, 95% CI 1.01–1.37, p = 0.04). Moderate-intensity PA was likely to have a slight protective effect (RR = 0.76, 95% CI 0.58–1.00, p = 0.05).
Conclusion
PA is associated with lower DR risk, and more studies should focus on the causality between them.
Get started with MATLAB for deep learning and AI with this in-depth primer. In this book, you start with machine learning fundamentals, then move on to neural networks, deep learning, and then convolutional neural networks. In a blend of fundamentals and applications, MATLAB Deep Learning employs MATLAB as the underlying programming language and tool for the examples and case studies in this book.
With this book, you'll be able to tackle some of today's real world big data, smart bots, and other complex data problems. You’ll see how deep learning is a complex and more intelligent aspect of machine learning for modern smart data analysis and usage.
You will:
• Use MATLAB for deep learning
• Discover neural networks and multi-layer neural networks
• Work with convolution and pooling layers
• Build a MNIST example with these layers
The availability of medical imaging data from clinical archives, research literature, and clinical manuals, coupled with recent advances in computer vision offer the opportunity for image-based diagnosis, teaching, and biomedical research. However, the content and semantics of an image can vary depending on its modality and as such the identification of image modality is an important preliminary step. The key challenge for automatically classifying the modality of a medical image is due to the visual characteristics of different modalities: some are visually distinct while others may have only subtle differences. This challenge is compounded by variations in the appearance of images based on the diseases depicted and a lack of sufficient training data for some modalities. In this paper, we introduce a new method for classifying medical images that uses an ensemble of different convolutional neural network (CNN) architectures. CNNs are a state-of-the-art image classification technique that learns the optimal image features for a given classification task. We hypothesise that different CNN architectures learn different levels of semantic image representation and thus an ensemble of CNNs will enable higher quality features to be extracted. Our method develops a new feature extractor by fine-tuning CNNs that have been initialised on a large dataset of natural images. The fine-tuning process leverages the generic image features from natural images that are fundamental for all images and optimises them for the variety of medical imaging modalities. These features are used to train numerous multi-class classifiers whose posterior probabilities are fused to predict the modalities of unseen images. Our experiments on the ImageCLEF 2016 medical image public dataset (30 modalities; 6776 training images and 4166 test images) show that our ensemble of fine-tuned CNNs achieves a higher accuracy than established CNNs. Our ensemble also achieves a higher accuracy than methods in the literature evaluated on the same benchmark dataset and is only overtaken by those methods that source additional training data.
Convolutional networks are at the core of most stateof-the-art computer vision solutions for a wide variety of tasks. Since 2014 very deep convolutional networks started to become mainstream, yielding substantial gains in various benchmarks. Although increased model size and computational cost tend to translate to immediate quality gains for most tasks (as long as enough labeled data is provided for training), computational efficiency and low parameter count are still enabling factors for various use cases such as mobile vision and big-data scenarios. Here we are exploring ways to scale up networks in ways that aim at utilizing the added computation as efficiently as possible by suitably factorized convolutions and aggressive regularization. We benchmark our methods on the ILSVRC 2012 classification challenge validation set demonstrate substantial gains over the state of the art: 21.2% top-1 and 5.6% top-5 error for single frame evaluation using a network with a computational cost of 5 billion multiply-adds per inference and with using less than 25 million parameters. With an ensemble of 4 models and multi-crop evaluation, we report 3.5% top-5 error and 17.3% top-1 error.
Very deep convolutional networks have been central to the largest advances in image recognition performance in recent years. One example is the Inception architecture that has been shown to achieve very good performance at relatively low computational cost. Recently, the introduction of residual connections in conjunction with a more traditional architecture has yielded state-of-the-art performance in the 2015 ILSVRC challenge; its performance was similar to the latest generation Inception-v3 network. This raises the question of whether there are any benefit in combining the Inception architecture with residual connections. Here we give clear empirical evidence that training with residual connections accelerates the training of Inception networks significantly. There is also some evidence of residual Inception networks outperforming similarly expensive Inception networks without residual connections by a thin margin. We also present several new streamlined architectures for both residual and non-residual Inception networks. These variations improve the single-frame recognition performance on the ILSVRC 2012 classification task significantly. We further demonstrate how proper activation scaling stabilizes the training of very wide residual Inception networks. With an ensemble of three residual and one Inception-v4, we achieve 3.08 percent top-5 error on the test set of the ImageNet classification (CLS) challenge
The Losartan Intervention For Endpoint (LIFE) reduction in hypertension study is a double-blind, prospective, parallel-group study comparing the effects of losartan with those of atenolol on the reduction of cardiovascular complications in patients (n 9,194) with essential hypertension and with electrocardiographically (ECG) documented left ventricular hypertrophy (LVH). Baseline blood pressure was 174.4/97.8 mm Hg (mean), age 66.9 years, body mass index 28.0 kg/m²; 54.1% were women and 12.5% had diabetes mellitus. This population will be treated until at least 1040 have a primary endpoint. After five scheduled visits and 12 months of follow-up, blood pressure decreased by 23.9/12.8 mm Hg to 150.5/85.1 mm Hg (target < 140/90 mm Hg). The mandatory titration level of ≤ 160/95 mm Hg was reached by 72.1% of the patients. At the 12-month visit, 22.7% of all patients were taking blinded study drug alone, 44.3% were taking blinded drug plus hydrochlorothiazide (HCTZ), and 17.7% were taking blinded drugs plus HCTZ and additional drugs. Controlling for all other variables, patients in the US received more medication and had 2.4 times the odds of achieving blood pressure control than patients in the rest of the study (P< .001). Previously untreated patients (n = 2530) had a larger initial decrease in blood pressure compared with those previously treated. Diabetics (n = 1148) needed more medication than nondiabetics to gain blood pressure control. Only 13.9% of the patients had discontinued blinded study drug and 1.4% missed the revisit at 12 months.
These data demonstrate both the successful lowering of blood pressure during 12 months of follow-up in a large cohort of patients with hypertension and LVH on ECG, but also emphasize the need for two or more drugs to control high blood pressure in most of these patients. Being previously treated and having diabetes were associated with less blood pressure response, whereas living in the US indicated better blood pressure control. It has been possible to keep most of these patients with complicated hypertension taking blinded study drug for 12 months.
Am J Hypertens 2000;13:899–906 © 2000 American Journal of Hypertension, Ltd.