Identification and Detecting COVID-19 from X-Ray Images Using Convolutional Neural Networks

The sudden increase of COVID-19 patients is alarming, and it requires quick diagnosis in a quick time. PCR testing is one of the most used methods to test and diagnose COVID, which is time-consuming. In this paper, we present an end-to-end technique that can detect COVID-19 using chest X-ray scans. We have trained and optimized a convolutional neural network (ConvNet), which was trained on a large COVID-19 dataset. We have performed a series of experiments on a number of different architectures. We have chosen the best performing network architecture and then carried on a series of additional experiments to find the optimal set of hyper-parameters and show and justify a number of data augmentation strategies that have allowed us to enhance our performance on the test set greatly. Our final trained ConvNet has managed to obtain a test accuracy of 97.89%. This high accuracy and very fast test speed can be beneficial to get quick COVID test results for further treatment. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Flu-Net: two-stream deep heterogeneous network to detect flu like symptoms from videos using grey wolf optimization algorithm.

The outbreak of COVID-19 (also known as Coronavirus) has put the entire world at risk. The disease first appears in Wuhan, China, and later spread to other countries, taking a form of a pandemic. In this paper, we try to build an artificial intelligence (AI) powered framework called Flu-Net to identify flu-like symptoms (which is also an important symptom of Covid-19) in people, and limit the spread of infection. Our approach is based on the application of human action recognition in surveillance systems, where videos captured by closed-circuit television (CCTV) cameras are processed through state-of-the-art deep learning techniques to recognize different activities like coughing, sneezing, etc. The proposed framework has three major steps. First, to suppress irrelevant background details in an input video, a frame difference operation is performed to extract foreground motion information. Second, a two-stream heterogeneous network based on 2D and 3D Convolutional Neural Networks (ConvNets) is trained using the RGB frame differences. And third, the features extracted from both the streams are combined using Grey Wolf Optimization (GWO) based feature selection technique. The experiments conducted on BII Sneeze-Cough (BIISC) video dataset show that our framework can 70% accuracy, outperforming the baseline results by more than 8%.

Deep Learning-Based Model for Face Mask Detection in the Era of COVID-19 Pandemic

Recent advancements in the growth of classification tasks and deep learning have culminated in the worldwide success of numerous practical applications. With the onset of COVID-19 pandemic, it becomes very important to use technology to help us control the infectious nature of the virus. Deep learning and image classification can help us detect face mask from a crowd of people. However, choosing the correct deep learning architecture can be crucial in the success of such an idea. This study presents a model for extracting features from face masks utilizing pre-trained models ConvNet, InceptionV3, MobileNet, DenseNet, ResNet50, and VGG19, as well as stacking a fully connected layer to solve the issue. On the face mask 12 k dataset, the study assesses the effectiveness of the suggested deep learning approaches for the task of facemask detection. The performance metrics used for analysis are loss, accuracy, validation loss, and validation accuracy. The maximum accuracy is achieved by DenseNet and MobileNet. Both the models gave a comparable and good accuracies in terms of training and validation (99.89% and 99.79%), respectively. Further, the paper also demonstrates the deployment of deep learning architecture in the real-world using Raspberry Pi 2B (1 GB RAM). © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Detection-Accuracy Enhancement of COVID-19 from Multi-Class Lung Diseases by Instrumenting CLAHE Integrated Deep Learning Technique

In this article we have investigated a detection-accuracy enhancement technique of COVID-19 from multi-class lung diseases by instrumenting the CLAHE integrated deep learning technique. The image population is distributed upon disjoint sets of the patients suffering from diseases like (a) Viral-Pneumonia, (b) Lung Opacity, (c) COVID-19 and (d) Normal persons. We have used the CLAHE algorithm to pre-process those images and incorporated the ConvNet algorithm with the help of the transfer learning strategy to perform the analysis. The study reveals that the processing of the X-ray images by CLAHE technique followed by ConvNet algorithm can enhance the performance of the above mentioned four exclusive classes of lung images. Keeping the X-ray image processing using the CLAHE technique intact, we have further explored the diagnosis methods by using several CNN models such as, InceptionResNetV2, InceptionV3 and DenseNet121. In our study, out of 2400 chest X-ray images we have used 80% for the training and 20% for the validation purpose. The comparative study of the performance matrices explicitly shows the enhancement in multi-class detection accuracy. The study also shows that the CLAHE integrated DenseNet121 model provides the best performance exhibiting a maximum accuracy of 98.33% for detecting COVID-19 from multiple diseases. Also we have compared the performance of the present technique with the earlier reported approaches. © 2022 IEEE.

COVID-19 Radiography Using ConvNets

The COVID-19 pandemic continues to have a negative impact on the fitness and well being of the worldwide population. A vital step in tackling the COVID-19 is a successful screening of patients, with one of the key screening approaches being radiological imaging using chest radiography. This study aims to automatically identify patients with COVID-19 pneumonia using digital x-ray images of the chest while increasing the accuracy of the diagnosis using Convolution Neural networks (CNN). The data-set consists of 5380 X-ray images consisting of 1345 X-ray images each of COVID patients, Lung Opacity, Normal patients and Viral Pneumonia. In this study, CNN based model have been proposed for the detection of coronavirus pneumonia infected patients using chest X-ray radiography and gives a classification accuracy of 93.77% (training accuracy of 99.81% and validation accuracy of 95.45%). © 2022 IEEE.

COVID-19 Diagnosis from Cough Acoustics using ConvNets and Data Augmentation

With the periodic rise and fall of COVID-19 and countries being inflicted by its waves, an efficient, economic, and effortless diagnosis procedure for the virus has been the utmost need of the hour. Amongst the infected subjects, the asymptomatic ones need not be entirely free of symptoms caused by the virus. They might not show any observable symptoms like the symptomatic subjects, but they may differ from uninfected ones in the way they cough. These differences in the coughing sounds are minute and indiscernible to the human ear, however, these can be captured using machine learning models. In this paper, we present a deep learning approach to analyze the acoustic dataset provided in Track 1 of the DiCOVA 2021 Challenge containing cough sound recordings belonging to both COVID-19 positive and negative examples. To perform the classification we propose a ConvNet model. It achieved an AUC score percentage of 72.23 on a blind test set provided in the challenge for an unbiased evaluation of the models. Moreover, the ConvNet model incorporated with Data Augmentation further increased the AUC score percentage from 72.23 to 87.07. It also outperformed the DiCOVA 2021 Challenge's baseline model by 23% thus, claiming the top position on the DiCOVA 2021 Challenge leaderboard. This paper proposes the use of Mel Frequency Cepstral Coefficients as the input features to the proposed model. © 2021 IEEE.

CNN Networks to Classify Cardiopulmonary Signals

At the moment, the world lives in a pandemic situation of COVID-19 and related variants, driving urgent needs for expanded assessments. A complementary support of related healthcare can be based on an intelligent system that can diagnose early onset of respiratory disorders. The convolutional neural networks (CNN) were implemented utilizing image data, reflecting bidimensional signals. Specifically, CNN has shown to be powerful tool in the context of cardiopulmonary sounds evaluation. The configurations of CNN contain convolutional layers to extract feature maps and fully connected layers to classify indicators of interest. Even though, learning algorithms use parameters like learning rate which can determine and attain CNN configuration less complex, with excellent results as reflected in the experiments we carried out, and which focused on achieved configuration of CNN with excellent results classifying heart sounds (HS) and lung sounds (LS).

Artificial intelligence enabled non-invasive T-ray imaging technique for early detection of coronavirus infected patients.

A new artificial intelligence (AI) supported T-Ray imaging system designed and implemented for non-invasive and non-ionizing screening for coronavirus-affected patients. The new system has the potential to replace the standard conventional X-Ray based imaging modality of virus detection. This research article reports the development of solid state room temperature terahertz source for thermograph study. Exposure time and radiation energy are optimized through several real-time experiments. During its incubation period, Coronavirus stays within the cell of the upper respiratory tract and its presence often causes an increased level of blood supply to the virus-affected cells/inter-cellular region that results in a localized increase of water content in those cells & tissues in comparison to its neighbouring normal cells. Under THz-radiation exposure, the incident energy gets absorbed more in virus-affected cells/inter-cellular region and gets heated; thus, the sharp temperature gradient is observed in the corresponding thermograph study. Additionally, structural changes in virus-affected zones make a significant contribution in getting better contrast in thermographs. Considering the effectiveness of the Artificial Intelligence (AI) analysis tool in various medical diagnoses, the authors have employed an explainable AI-assisted methodology to correctly identify and mark the affected pulmonary region for the developed imaging technique and thus validate the model. This AI-enabled non-ionizing THz-thermography method is expected to address the voids in early COVID diagnosis, at the onset of infection.

Diagnosing COVID-19 From Chest CT Scan Images Using Deep Learning Models

A novel coronavirus named COVID-19 has spread speedily and has triggered a worldwide outbreak of respiratory illness. Early diagnosis is always crucial for pandemic control. Compared to RT-PCR, chest computed tomography (CT) imaging is the more consistent, concrete, and prompt method to identify COVID-19 patients. For clinical diagnostics, the information received from computed tomography scans is critical. So there is a need to develop an image analysis technique for detecting viral epidemics from computed tomography scan pictures. Using DenseNet, ResNet, CapsNet, and 3D-ConvNet, four deep machine learning-based architectures have been proposed for COVID-19 diagnosis from chest computed tomography scans. From the experimental results, it is found that all the architectures are providing effective accuracy, of which the COVID-DNet model has reached the highest accuracy of 99%. Proposed architectures are accessible at https://github.com/shamiktiwari/CTscanCovi19 can be utilized to support radiologists and reserachers in validating their initial screening.

Stacked Dark COVID-Net: A Multi-class Multi-label Classification Approach for Diagnosing COVID-19 Using Chest X-Ray Images

Most challenging yet, the need of the hour is accurate diagnosis of COVID-19, as the Coronavirus cases are increasing drastically day-by-day. Ceaseless efforts by the researchers and innovators have led to the development of several diagnostic models based on Deep Learning for effective diagnosis of COVID-19. However, the Deep Learning techniques that have been developed so far, fail to address major challenges such as overfitting, stability, computation overhead due to the usage of the massive volume of parameters and problems associated with the multi-class classification. Also in the medical perspective, researchers often suffer to identify the infinitesimal difference that exists in the radiographic images among the several lung diseases which makes the decision-making process difficult. Thus, to curb the crisis and to provide promising solutions & expertise for accurate diagnosis, this paper presents a novel lightweight multi-class multi-label COVID-19 detection model to assist physicians with greater ease to fight against this pandemic situation. Radiographic images are pre-processed using Contrast Limited Adaptive Histogram Equalization (CLAHE) and classified using novel Stacked Dark COVID-Net. The proposed model is validated using chest X-ray images and the results confirm the efficacy of the proposed model in terms of classification accuracy, sensitivity, specificity and stability. © 2022, Springer Nature Switzerland AG.

Video-Based Student Engagement Estimation via Time Convolution Neural Networks for Remote Learning

Given the recent outbreak of COVID-19 pandemic globally, most of the schools and universities have adapted many of the learning materials and lectures to be delivered online. As a result, the necessity to have some quantifiable measures of how the students are perceiving and interacting with this ‘new normal’ way of education became inevitable. In this work, we are focusing on the engagement metric which was shown in the literature to be a strong indicator of how students are dealing with the information and the knowledge being presented to them. In this regard, we have proposed a novel data-driven approach based on a special variant of convolutional neural networks that can predict the students’ engagement levels from a video feed of students’ faces. Our proposed framework has achieved a promising mean-squared error (MSE) score of only 0.07 when evaluated on a real dataset of students taking an online course. Moreover, the proposed framework has achieved superior results when compared with two baseline models that are commonly utilised in the literature for tackling this problem. © 2022, Springer Nature Switzerland AG.