Deep learning-based COVID-19 triage tool: An observational study on an X-ray dataset

Background: Easy availability, low cost, and low radiation exposure make chest radiography an ideal modality for coronavirus disease 2019 (COVID-19) detection. Objective(s): In this study, we propose the use of an artificial intelligence (AI) algorithm to automatically detect abnormalities associated with COVID-19 on chest radiographs. We aimed to evaluate the performance of the algorithm against the interpretation of radiologists to assess its utility as a COVID-19 triage tool. Material(s) and Method(s): The study was conducted in collaboration with Kaushalya Medical Trust Foundation Hospital, Thane, Maharashtra, between July and August 2020. We used a collection of public and private datasets to train our AI models. Specificity and sensitivity measures were used to assess the performance of the AI algorithm by comparing AI and radiology predictions using the result of the reverse transcriptase-polymerase chain reaction as reference. We also compared the existing open-source AI algorithms with our method using our private dataset to ascertain the reliability of our algorithm. Result(s): We evaluated 611 scans for semantic and non-semantic features. Our algorithm showed a sensitivity of 77.7% and a specificity of 75.4%. Our AI algorithm performed better than the radiologists who showed a sensitivity of 75.9% and specificity of 75.4%. The open-source model on the same dataset showed a large disparity in performance measures with a specificity of 46.5% and sensitivity of 91.8%, thus confirming the reliability of our approach. Conclusion(s): Our AI algorithm can aid radiologists in confirming the findings of COVID-19 pneumonia on chest radiography and identifying additional abnormalities and can be used as an assistive and complementary first-line COVID-19 triage tool.Copyright © Cancer Research, Statistics, and Treatment.

Comparative study of Deep Learning Models for Covid 19 Diagnosis

The Covid 19 pandemic that started a couple of years ago has had a devastating effect on mankind across the globe. The disease had no known treatment. Early detection and prevention was very important to curtail the effects of the Pandemic. In this work two deep learning models the RestNet and the models are proposed for diagnosing Corona from chest X-rays and CT scans. The models were trained with publicly available data sets of covid and non covid images. It has been found that Inception V3 performs better than ResNet for chest x-rays and RestNet performs better for CT Scans. The performance of the RestNet is found to be similar for both the chest x-rays and CT scans datasets. © 2023 IEEE.

LWSNet - a novel deep-learning architecture to segregate Covid-19 and pneumonia from x-ray imagery.

Automatic detection of lung diseases using AI-based tools became very much necessary to handle the huge number of cases occurring across the globe and support the doctors. This paper proposed a novel deep learning architecture named LWSNet (Light Weight Stacking Network) to separate Covid-19, cold pneumonia, and normal chest x-ray images. This framework is based on single, double, triple, and quadruple stack mechanisms to address the above-mentioned tri-class problem. In this framework, a truncated version of standard deep learning models and a lightweight CNN model was considered to conviniently deploy in resource-constraint devices. An evaluation was conducted on three publicly available datasets alongwith their combination. We received 97.28%, 96.50%, 97.41%, and 98.54% highest classification accuracies using quadruple stack. On further investigation, we found, using LWSNet, the average accuracy got improved from individual model to quadruple model by 2.31%, 2.55%, 2.88%, and 2.26% on four respective datasets.

Deep Learning Convolutional Neural Network for SARS-CoV-2 Detection Using Chest X-Ray Images

The COVID-19 coronavirus illness is caused by a newly discovered species of coronavirus known as SARS-CoV-2. Since COVID-19 has now expanded across many nations, the World Health Organization (WHO) has designated it a pandemic. Reverse transcription-polymerase chain reaction (RT-PCR) is often used to screen samples of patients showing signs of COVID-19;however, this method is more expensive and takes at least 24 hours to get a positive or negative response. Thus, an immediate and precise method of diagnosis is needed. In this paper, chest X-rays will be utilized through a deep neural network (DNN), based on a convolutional neural network (CNN), to detect COVID-19 infection. Based on their X-rays, those with COVID-19 indications may be categorized as clean, infected with COVID-19 or suffering from pneumonia, according to the suggested CNN network. Sample pieces from every group are used in experiments, and categorization is performed by a CNN. While experimenting, the CNN-derived features were able to generate the maximum training accuracy of 94.82% and validation accuracy of 94.87%. The F1-scores were 97%, 90% and 96%, in clearly categorizing patients afflicted by COVID-19, normal and having pneumonia, respectively. Meanwhile, the recalls are 95%, 91% and 96% for COVID-19, normal and pneumonia, respectively. © 2023 by the author(s). Licensee Prague University of Economics and Business, Czech Republic.

Deep neural network for monitoring the growth of COVID-19 epidemic using meteorological covariates

Growth of an epidemic is influenced by the natural variation in climatic conditions and enforcement variation in government stringency policies. Though these variations do not prompt an instant change in the growth of an epidemic, effects of climatic conditions and stringency policies become apparent over time. Time-lagged relationships and functional dynamic connectivity among meteorological covariates and stringency levels generate many lagged features deemed to be important for prediction of reproduction rate, a measure for growth of an epidemic. This empirical study examines the importance scores of lagged features and implements distributed lag inspired feature selection with back testing for model selection and forecasting. A verification forecasting scheme is developed for continuous monitoring of the growth of an epidemic. We have demonstrated the monitoring process by computing a week ahead expected target of the reproduction rate and then by computing a one day ahead verification forecast to evaluate the progress towards the expected target. This evaluation procedure will aid the analysts with a decision making tool for any early adjustment of control options to suppress the transmission.

Deep Rank-Based Average Pooling Network for Covid-19 Recognition

(Aim) To make a more accurate and precise COVID-19 diagnosis system, this study proposed a novel deep rank-based average pooling network (DRAPNet) model, i.e., deep rank-based average pooling network, for COVID-19 recognition. (Methods) 521 subjects yield 1164 slice images via the slice level selection method. All the 1164 slice images comprise four categories: COVID-19 positive;community-acquired pneumonia;second pulmonary tuberculosis;and healthy control. Our method firstly introduced an improved multiple-way data augmentation. Secondly, an n-cony rank-based average pooling module (NRAPM) was proposed in which rank-based pooling-particularly, rank-based average pooling (RAP)-was employed to avoid overfitting. Third, a novel DRAPNet was proposed based on NRAPM and inspired by the VGG network. Grad-CAM was used to generate heatmaps and gave our AI model an explainable analysis. (Results) Our DRAPNet achieved a micro-averaged F1 score of 95.49% by 10 runs over the test set. The sensitivities of the four classes were 95.44%, 96.07%, 94.41%, and 96.07%, respectively. The precisions of four classes were 96.45%, 95.22%, 95.05%, and 95.28%, respectively. The F1 scores of the four classes were 95.94%, 95.64%, 94.73%, and 95.67%, respectively. Besides, the confusion matrix was given. (Conclusions) The DRAPNet is effective in diagnosing COVID-19 and other chest infectious diseases. The RAP gives better results than four other methods: strided convolution, l(2)-norm pooling, average pooling, and max pooling.

Susceptible exposed infectious recovered-machine learning for COVID-19 prediction in Saudi Arabia

Susceptible exposed infectious recovered (SEIR) is among the epidemiological models used in forecasting the spread of disease in large populations. SEIR is a fitting model for coronavirus disease (COVID-19) spread prediction. Somehow, in its original form, SEIR could not measure the impact of lockdowns. So, in the SEIR equations system utilized in this study, a variable was included to evaluate the impact of varying levels of social distance on the transmission of COVID-19. Additionally, we applied artificial intelligence utilizing the deep neural network machine learning (ML) technique. On the initial spread data for Saudi Arabia that were available up to June 25th, 2021, this improved SEIR model was used. The study shows possible infection to around 3.1 million persons without lockdown in Saudi Arabia at the peak of spread, which lasts for about 3 months beginning from the lockdown date (March 21st). On the other hand, the Kingdom's current partial lockdown policy was estimated to cut the estimated number of infections to 0.5 million over nine months. The data shows that stricter lockdowns may successfully flatten the COVID-19 graph curve in Saudi Arabia. We successfully predicted the COVID-19 epidemic's peaks and sizes using our modified deep neural network (DNN) and SEIR model. © 2023 Institute of Advanced Engineering and Science. All rights reserved.

Residential Load Forecasting based on Deep Neural Network

This paper presents residential load forecasting using multivariate multi-step Deep Neural Networks (DNN) such as LSTM, CNN, Stacked LSTM, and Hybrid CNN-LSTM. A preliminary Exploratory Data Analysis (EDA) is conducted, and the decision variables are identified. An elbowing method is used to determine the number of clusters. Data is categorized based on weekdays, weekends, vacations, and Covid-Lockdown. Dimensionality-reduction using principal component analysis (PCA) is conducted. Seasonality-based clustering is found to improve the DNN model prediction accuracy further. A comparative analysis employs error metrics such as RMSE, MSE, MAPE, and MAE. The multivariate LSTM model with feedback is found to be the best fit model with the better performance indices. © 2023 IEEE.

Design of an Intelligent Educational Evaluation System Using Deep Learning

Nowadays, online education has been a more general demand in context of COVID-19 epidemic. The intelligent educational evaluation systems assisted by intelligent techniques are in urgent demand. To deal with this issue, this paper introduces the strong information processing ability of deep learning, and proposes the design of an intelligent educational evaluation system using deep learning. Inside the algorithm part, the low-complexity offset minimal sum (OMS) is selected as the front-end processor of deep neural network, so as to reduce following computational complexity in deep neural network. And the deep neural network is adopted as the major calculation backbone. In this paper, our OMS deep neural network parameters are 23 and 57 compared with other parameters, which can save about 59.64% of the network parameters, and the training time is 11270 s and 25000 s respectively, which saves the training time 54.92%. It can be also reflected from experiments that the proposal further improves the performance of unbalanced data classification in this problem scenario. © 2013 IEEE.

Automated semantic lung segmentation in chest CT images using deep neural network.

Lung segmentation algorithms play a significant role in segmenting theinfected regions in the lungs. This work aims to develop a computationally efficient and robust deep learning model for lung segmentation using chest computed tomography (CT) images with DeepLabV3 + networks for two-class (background and lung field) and four-class (ground-glass opacities, background, consolidation, and lung field). In this work, we investigate the performance of the DeepLabV3 + network with five pretrained networks: Xception, ResNet-18, Inception-ResNet-v2, MobileNet-v2 and ResNet-50. A publicly available database for COVID-19 that contains 750 chest CT images and corresponding pixel-labeled images are used to develop the deep learning model. The segmentation performance has been assessed using five performance measures: Intersection of Union (IoU), Weighted IoU, Balance F1 score, pixel accu-racy, and global accuracy. The experimental results of this work confirm that the DeepLabV3 + network with ResNet-18 and a batch size of 8 have a higher performance for two-class segmentation. DeepLabV3 + network coupled with ResNet-50 and a batch size of 16 yielded better results for four-class segmentation compared to other pretrained networks. Besides, the ResNet with a fewer number of layers is highly adequate for developing a more robust lung segmentation network with lesser computational complexity compared to the conventional DeepLabV3 + network with Xception. This present work proposes a unified DeepLabV3 + network to delineate the two and four different regions automatically using CT images for CoVID-19 patients. Our developed automated segmented model can be further developed to be used as a clinical diagnosis system for CoVID-19 as well as assist clinicians in providing an accurate second opinion CoVID-19 diagnosis.

Design of ear-contactless stethoscope and improvement in the performance of deep learning based on CNN to classify the heart sound.

Cardiac-related disorders are rapidly growing throughout the world. Accurate classification of cardiovascular diseases is an important research topic in healthcare. During COVID-19, auscultating heart sounds was challenging as health workers and doctors wear protective clothing, and direct contact with patients can spread the outbreak. Thus, contactless auscultation of heart sound is necessary. In this paper, a low-cost ear contactless stethoscope is designed where auscultation is done with the help of a bluetooth-enabled micro speaker instead of an earpiece. The PCG recordings are further compared with other standard electronic stethoscopes like Littman 3 M. This work is made to improve the performance of deep learning-based classifiers like recurrent neural networks (RNN) and convolutional neural networks (CNN) for different valvular heart problems using tuning of hyperparameters like learning rate of optimizers, dropout rate, and hidden layer. Hyper-parameter tuning is used to optimize the performances of various deep learning models and their learning curves for real-time analysis. The acoustic, time, and frequency domain features are used in this research. The investigation is made on the heart sounds of normal and diseased patients available from the standard data repository to train the software models. The proposed CNN-based inception network model achieved an accuracy of 99.65 ± 0.06% on the test dataset with a sensitivity of 98.8 ± 0.05% and specificity of 98.2 ± 0.19%. The proposed hybrid CNN-RNN architecture attained 91.17 ± 0.03% accuracy on test data after hyperparameter optimization, whereas the LSTM-based RNN model achieved 82.32 ± 0.11% accuracy. Finally, the evaluated results were compared with machine learning algorithms, and the improved CNN-based Inception Net model is the most effective among others.

LogNNet model as a fast, simple and economical AI instrument in the diagnosis and prognosis of COVID-19.

Rapid and effective detection of the diagnosis and prognosis of COVID-19 disease is important in terms of reducing the mortality of the disease and reducing the pressure on health systems. Methods such as PCR testing and computed tomography used for this purpose in current health systems are costly, require an expert team and take time. This study offers a fast, economical and reliable approach for the early diagnosis and prognosis of infectious diseases, especially COVID-19. For this purpose, characteristics of a large population of COVID-19 patients were determined (51 different routine blood values) and calibrated. In order to determine the diagnosis and prognosis of the disease, the calibrated features were run with the LogNNet model. LogNNet has a simple algorithm and performance indicators comparable to the most efficient algorithms available.This approach pointed out that routine blood values contain important information, especially in the detection of COVID-19, and showed that the LogNNet model can be used as an economical, safe and fast alternative tool in the diagnosis of this disease.-In the LogNNet feedforward neural network, a feature vector is passed through a specially designed reservoir matrix and transformed into a new feature vector of a different size, increasing the classification accuracy.-The presented network architecture can achieve 80%-99% classification accuracy using a range of weightings on devices with a total memory size of 1 to 29 kB constrained.-Due to the chaotic mapping procedures, the RAM usage in the LogNNet neural network processing process is greatly reduced. Hence, optimization of chaotic map parameters has an important function in LogNNet neural network application.

Drowsiness Detection System for Masked Face Based on Deep Neural Network and Haar Cascade

Nowadays, technology is growing rapidly followed by modernization. Face detection technology is one technology that has been developed and applied in various sectors such as biometrics recognition systems, retrieval systems, database indexing in digital video, security systems with restricted area access control, video conferencing, and human interaction systems. Eye detection is a further development of face detection in which the image of a human face was detected to be processed by detecting the location of both eyes on the face. Nowadays, the eye detection system can be used as a means of developing more complex applications and can be applied directly in the aspect of technology that uses eye detection like, eye state detection system, drowsiness and fatigue detection system, safety driving support systems or driver assistance system. In this study we propose drowsiness detection system utilizing current novel classification model such as Deep Neural Network (DNN), combined with Haar Cascade. The DNN is utilized to detect face, while Haar Cascade is utilized for detecting the eyes and its state on the detected face. In this study, due to Covid19 pandemic, we focused on developing the classifiers for detecting the face with mask. Apart from that, our proposed classifiers are also capable of identifying non-masked faces. The experimental result showed that by utilizing DNN and Haar Cascade, our proposed system could reach accuracy, precision, recall, and f1 measure as much as 81%, 88%, 80%, and 84%, respectively. © 2022 IEEE.

Identification and Localization COVID-19 Abnormalities on Chest Radiographs

Solutions to screen and diagnose positive patients for the SARS-CoV-2 promptly and efficiently are critical in the context of the COVID-19 pandemic's complex evolution. Recent researches have demonstrated the efficiency of deep learning and particularly convolutional neural networks (CNNs) in classifying and detecting lung disease-related lesions from radiographs. This paper presents a solution using ensemble learning techniques on advanced CNNs to classify as well as localize COVID-19-related abnormalities in radiographs. Two classifiers including EfficientNetV2 and NFNet are combined with three detectors, DETR, Yolov7 and EfficientDet. Along with gathering and training the model on a large number of datasets, image augmentation and cross validation are also addressed. Since then, this study has shown promising results and has received excellent marks in the Society for Imaging Informatics in Medicine's competition. The analysis in model selection for the trade-off between speed and accuracy is also given. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Design of an Efficient Approach for Performance Enhancement of COVID-19 Detection Using Auxiliary GoogLeNet by Using Chest CT Scan Images

In every country on this planet, COVID-19 disease s right now one of the most unsafe issues. The expedient and precise space of the Covid virus infection s major to see and take better treatment for the infected patients will increase the chance of saving their lives. The quick spread of the Covid virus has blended complete interest and caused greater than 10 lacks cases to date. To battle this spread, Chest CTs arise as a basic demonstrative contraption for the clinical association of COVID-19 related to a lung illness. A modified confirmation device is essential for assisting in the screening for COVID-19 pneumonia by making use of chest CT imaging. The COVID-19 illness detection utilizing supplementary GoogLeNet is shown in this study. Deep Convolutional Neural Networks were built by researchers at Google, and one of their innovations was the Inception Network. GoogLeNet is a 22-layer deep convolutional neural network that is a variation of the inception Network. GoogLeNet is utilized for a variety of additional computer vision applications nowadays, including face identification and recognition, adversarial training, and so on. The findings indicate that the GoogLeNet method is superior to the CNN Method in terms of its ability to detect COVID-19 sickness. © 2022 IEEE.

AIDrugApp: artificial intelligence-based Web-App for virtual screening of inhibitors against SARS-COV-2

Currently, there is no effective cure for SARS-COVID-19 diseases. The identification of novel therapeutic targets and drug-like compounds is required for the development of anti-COVID-19 drugs. Virtual screening is currently the most significant component for identifying drug-like molecules from large datasets for drug design and development. However, there are no effective easily available and user-friendly applications for virtual screening of drug leads against SARS-COV-2. Therefore, we have developed a user-friendly web-app named ‘AIDrugApp' for the virtual screening of inhibitor molecules against SARS-CoV-2. AIDrugApp is a novel open-access, deep learning AI-based inhibitory activity prediction and data statistics visualisation platform. Users can predict the inhibitory activities (Active/Inactive) and pIC-50 values of new compounds against SARS-CoV-2 replicase polyprotein, 3CLpro and human angiotensin-converting enzymes. It is also useful for virtual screening of chemical features of molecules towards SARS-COVID-19 clinical trial bioactivities. This paper presents the development and architecture of AIDrugApp. We also present two case studies where large sets of molecules were screened using the ‘Bioactivity Prediction' module of our app. Screened molecules were analysed further for validation by molecular docking and ADME analysis to identify the potential drug candidates.

Application of Artificial Intelligence on Post Pandemic Situation and Lesson Learn for Future Prospects

Coronavirus disease (COVID-19) pandemic has intensively damaged human socio-economic lives and the growth of countries around the world. Many efforts have been made in the direction of artificial intelligence (AI) techniques to detect the corona at an early stage and take necessary precautions to stop it from spreading or recovery from the infection. However, the situation and solutions are still challenging. In this paper, we proposed various technological aspects, solutions using a supervised/unsupervised manner and continuous health monitoring with physiological parameters. Finally, the performance of COVID-19 detection with Gaussian mixture model-universal background model (GMM-UBM) technique using the voice signal has been demonstrated. The developed system achieves the COVID-19 detection performance in terms of areas under receiver operating characteristic (ROC) curves in the range 60–67%. Moreover, the various lessons learned from the current COVID-19 crisis are presented for future directions.

Image Processing and Deep Neural Networks for Face Mask Detection

The proper use of a mask is crucial for lowering COVID 19 and transmission. According to the research, transmission is completely decreased when the mask is used appropriately. Factors like sunlight and several items can affect how appropriatel y applied face masks are classified and detected. Cotton masks, sponge masks, scarves, and other options greatly lessen the effect of personal protection in such circumstances. The research suggests a novel modified formula for classifying masks into three categories—a proper mask, a no mask, and an erroneous mask—using deep learning and machine learning. First, we provide a brand-new face mask classification and detection algorithm that combines deep learning, the viola Jones method, and Efficient-Yolov3 Wearing a mask, not wearing a mask, or wearing the wrong mask are the three options. On the dataset with or without mask pictures, the suggested system outperforms and is more accurate when compared to existing techniques. The results of experiments and analysis are also based on the classification knowledge set. In comparison to the present methodology's categorization accuracy of 84%, the anticipated formula boosted it to 97%. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Feature Extraction Based Face Mask Detection

Face masks are becoming an essential part of every person's life to protect against COVID-19, air pollution, etc. In public venues like airports, hospitals, malls, and many other locations, face masks are now required. This research introduces a unique method for face mask identification that combines deep learning and machine learning. Thepade Sorted 10-ary Block Trucation Code with RGB colour plane, Thepade Sorted Block Truncation Code with LUV colour plane, Local Binary Pattern method, and Gray Level Co-occurrence Matrics are used for feature extraction. Deep learning techniques are used to train these retrieved characteristics. Accuracy, precision, and f1-score are the performance measures used to evaluate performance. © 2022 IEEE.

Customized EfficientNet for Histopathology Image Representation

In recent years, and amplified by the COVID-19 pandemic, the digitization of pathology has gained a considerable attention. Digital pathology provides crucial advantages compared to conventional light microscopy, including more efficient workflows, more accurate diagnosis and treatment planning, and easier collaboration. Despite promising progress, there are some critical challenges related to classifying images in digital pathology, such as huge input sizes (e.g., gigapixels) and expensive processing time. Most of the existing models for classification of histopathology images are very large and accordingly have many parameters to be learned/optimized. In addition, due to the large size of Whole Slide Images (WSIs), e.g., 100,000 × 100,000 pixels, models require enormous computational times to achieve reliable results. In order to address these challenges, we propose a more compact network which is customized to classify cancer subtypes with lower computation time and memory complexity. This model is based on EfficientNet topology, but it is tailored for classifying histopathology images. The utilized model is evaluated over three tumor types brain, lung, and kidney from The Cancer Genome Atlas (TCGA) public repository. Since the pre-trained EfficientNet works properly with the specific size of images, an effective approach is proposed to adjust the size of input images. The proposed model can be trained with a much smaller training set for applications such as image search that require robust and compact representations. The results show that the proposed model, compared to state-of-the-art models, i.e., KimiaNet, can classify cancer subtypes more accurately and provides superior results. In addition, the proposed model achieves memory and computational efficiency in the training phase and is a more compact deep topology compared to KimiaNet. © 2022 IEEE.