Performance, Trust, or both? COVID-19 Diagnosis and Prognosis using Deep Ensemble Transfer Learning on X-ray Imagesg±

The COVID-19 pandemic still affects most parts of the world today. Despite a lot of research on diagnosis, prognosis, and treatment, a big challenge today is the limited number of expert radiologists who provide diagnosis and prognosis on X-Ray images. Thus, to make the diagnosis of COVID-19 accessible and quicker, several researchers have proposed deep-learning-based Artificial Intelligence (AI) models. While most of these proposed machine and deep learning models work in theory, they may not find acceptance among the medical community for clinical use due to weak statistical validation. For this article, radiologists' views were considered to understand the correlation between the theoretical findings and real-life observations. The article explores Convolutional Neural Network (CNN) classification models to build a four-class viz. "COVID-19", "Lung Opacity", "Pneumonia", and "Normal"classifiers, which also provide the uncertainty measure associated with each class. The authors also employ various pre-processing techniques to enhance the X-Ray images for specific features. To address the issues of over-fitting while training, as well as to address the class imbalance problem in our dataset, we use Monte Carlo dropout and Focal Loss respectively. Finally, we provide a comparative analysis of the following classification models - ResNet-18, VGG-19, ResNet-152, MobileNet-V2, Inception-V3, and EfficientNet-V2, where we match the state-of-the-art results on the Open Benchmark Chest X-ray datasets, with a sensitivity of 0.9954, specificity of 0.9886, the precision of 0.9880, F1-score of 0.9851, accuracy of 0.9816, and receiver operating characteristic (ROC) of the area under the curve (AUC) of 0.9781 (ROC-AUC score). © 2022 ACM.

Research on Chest X-Ray Image Diagnosis of Covid-19 Based on Improved ResNet

With the outbreak of covid-19 in 2020, timely and effective diagnosis and treatment of each covid-19 patient is particularly important. This paper combines the advantages of deep learning in image recognition, takes RESNET as the basic network framework, and carries out the experiment of improving the residual structure on this basis. It is tested on the open source new coronal chest radiograph data set, and the accuracy rate is 82.3%. Through a series of experiments, the training model has the advantages of good generalization, high accuracy and fast convergence. This paper proves the feasibility of the improved residual neural network in the diagnosis of covid-19. © 2023 SPIE.

Multi-Joint Unit Network for COVID-19 Diagnosis using X-Ray and CT Images

The global COVID-19 pandemic has caused a health crisis globally. Automated diagnostic methods can control the spread of the pandemic, as well as assists physicians to tackle high workload conditions through the quick treatment of affected patients. Owing to the scarcity of medical images and from different resources, the present image heterogeneity has raised challenges for achieving effective approaches to network training and effectively learning robust features. We propose a multi-joint unit network for the diagnosis of COVID-19 using the joint unit module, which leverages the receptive fields from multiple resolutions for learning rich representations. Existing approaches usually employ a large number of layers to learn the features, which consequently requires more computational power and increases the network complexity. To compensate, our joint unit module extracts low-, same-, and high-resolution feature maps simultaneously using different phases. Later, these learned feature maps are fused and utilized for classification layers. We observed that our model helps to learn sufficient information for classification without a performance loss and with faster convergence. We used three public benchmark datasets to demonstrate the performance of our network. Our proposed network consistently outperforms existing state-of-the-art approaches by demonstrating better accuracy, sensitivity, and specificity and F1-score across all datasets. © 2022 ACM.

Deep-learning models for Covid-19 Detection Using Chest X-Ray Images

Deep Convolutional Neural Networks are a form of neural network that can categorize, recognize, or separate images. The problem of COVID-19 detection has become the world's most complex challenge since 2019. In this research work, Chest X-Ray images are used to detect patients' Covid Positive or Negative with the help of pre-trained models: VGG16, InceptionV3, ResNet50, and InceptionResNetV2. In this paper, 821 samples are used for training, 186 samples for validation, and 184 samples are used for testing. Hybrid model InceptionResNetV2 has achieved overall maximum accuracy of 94.56% with a Recall value of 96% for normal CXR images, and a precision of 95.12% for Covid Positive images. The lowest accuracy was achieved by the ResNet50 model of 92.93% on the testing dataset, and a Recall of 93.93% was achieved for the normal images. Throughout the implementation process, it was discovered that factors like epoch had a considerable impact on the model's accuracy. Consequently, it is advised that the model be trained with a sufficient number of epochs to provide reliable classification results. The study's findings suggest that deep learning models have an excellent potential for correctly identifying the covid positive or covid negative using CXR images. © 2023 IEEE.

Chest X-ray Imaging System for Early Detection of Tuberculosis

Today it is observed that few people respect the biosecurity measures announced by the WHO, which aimed to reduce the amount of COVID-19 infection among people, even knowing that this virus has not disappeared from our environment, being an unprecedented infection in the world. It should be noted that before this pandemic, tuberculosis affected millions of people, having a great role because it is highly contagious and directly affects the lungs, although it has a cure, if it is not treated in time it can be fatal for the person, although there are many methods of detection of tuberculosis, one that is most often used is the diagnosis by chest x-ray, although it has low specificity, when the image processing technique is applied, tuberculosis would be accurately detected. In view of this problem, in this article a chest X-ray image processing system was conducted for the early detection of tuberculosis, helping doctors to detect tuberculosis accurately and quickly by having a second opinion by the system in the analysis of the chest x-ray, prevents fatal infections in patients. Through the development of the tuberculosis early detection system, it was possible to observe the correct functioning of the system with an efficiency of 97.84% in the detection of tuberculosis, detailing the characteristics presented by normal or abnormal images so that the doctor detects tuberculosis in the patient early. © 2023 IEEE.

Detection and Classification of Pneumonia Using Deep Learning by the Dense Net-121 Model

A chronic respiratory disease known as pneumonia can be devastating if it is not identified and treated in a timely manner. For successful treatment and better patient outcomes, pneumonia must be identified early and properly classified. Deep learning has recently demonstrated considerable promise in the area of medical imaging and has successfully applied for a few image-based diagnosis tasks, including the identification and classification of pneumonia. Pneumonia is a respiratory illness that produces pleural effusion (a condition in which fluids flood the lungs). COVID-19 is becoming the major cause of the global rise in pneumonia cases. Early detection of this disease provides curative therapy and increases the likelihood of survival. CXR (Chest X-ray) imaging is a common method of detecting and diagnosing pneumonia. Examining chest X-rays is a difficult undertaking that often results in variances and inaccuracies. In this study, we created an automatic pneumonia diagnosis method, also known as a CAD (Computer-Aided Diagnosis), which may significantly reduce the time and cost of collecting CXR imaging data. This paper uses deep learning which has the potential to revolutionize in the area of medical imaging and has shown promising results in the detection and classification of pneumonia. Further research and development in this area is needed to improve the accuracy and reliability of these models and make them more accessible to healthcare providers. These models can provide fast and accurate results, with high sensitivity and specificity in identifying pneumonia in chest X-rays. © 2023 IEEE.

X-rays imaging analysis for early diagnosis of the thoracic disorders using Capsule Neural Network with Transfer learning

Over the past few years, millions of people around the world have developed thoracic ailments. MRI, CT scan, reverse transcription, and other methods are among those used to detect thoracic disorders. These procedures demand medical knowledge and are exceedingly pricy and delicate. An alternate and more widely used method to diagnose diseases of the chest is X-ray imaging. The goal of this study was to increase detection precision in order to develop a computationally assisted diagnostic tool. Different diseases can be identified by combining radiological imaging with various artificial intelligence application approaches. In this study, transfer learning (TL) and capsule neural network techniques are used to propose a method for the automatic detection of various thoracic illnesses utilizing digitized chest X-ray pictures of suspected patients. Four public databases were combined to build a dataset for this purpose. Three pre trained convolutional neural networks (CNNs) were utilized in TL with augmentation as a preprocessing technique to train and evaluate the model. Pneumonia, COVID19, normal, and TB (Tb) were the four class classifiers used to train the network to categorize. © 2023 IEEE.

CovidXNet: An Ensemble of Deep Neural Networks for Non-invasive Binary Identification of COVID-19 through Chest Radiographs

This study compared five different image classification algorithms, namely VGG16, VGG19, AlexNet, DenseNet, and ConVNext, based on their ability to detect and classify COVID-19-related cases given chest X-ray images. Using performance metrics like accuracy, F1 score, precision, recall, and MCC compared these intelligent classification algorithms. Upon testing these algorithms, the accuracy for each model was quite unsatisfactory, ranging from 80.00% to 92.50%, provided it is for medical application. As such, an ensemble learning-based image classification model, made up of AlexNet and VGG19 called CovidXNet, was proposed to detect COVID-19 through chest X-ray images discriminating between health and pneumonic lung images. CovidXNet achieved an accuracy of 97.00%, which was significantly better considering past results. Further studies may be conducted to increase the accuracy, particularly for identifying and classifying chest radiographs for COVID-19-related cases, since the current model may still provide false negatives, which may be detrimental to the prevention of the spread of the virus. © 2022 IEEE.

Development of a Decentralized Web App (D-App) for Covid-19 using AI

A horrifying number of people died because of the COVID-19 pandemic. There was an unexpected threat to food systems, public health, and the workplace. The pandemic has severely disturbed society and there was a serious impediment to the economy. The world went through an unprecedented state of chaos during this period. To avoid anything similar, we can only be cautious. The project aims to develop a web application for the preliminary detection of COVID-19 using Artificial Intelligence(AI). This project would enable faster coordination, secured data storage, and normalized statistics. First, the available chest X-ray datasets were collected and classified as Covid, Non-Covid, and Normal. Then they were trained using various state-of-the-art pre-trained Convolutional Neural Network (CNN) models with the help of Tensor-flow. Further, they were ranked based on their accuracy. The best-performing models were ensembled into a single model to improve the performance. The model with the highest accuracy was transformed into an application programming interface (API) and integrated with the Decentralized application (D-App). The user needs to upload an image of their chest X-ray, and the D-App then suggests if they should take a reverse transcription-polymerase chain reaction (RT-PCR) test for confirmation. © 2022 IEEE.

Classifying Chest X-Ray Images using Residual Neural Network

This unprecedented time of the COVID-19 outbreak challenged the status-quo whether it is on business operation, political leadership, scientific capability, engineering implementation, data analysis, and strategic thinking, in terms of resiliency, agility, and innovativeness. Due to some identified constraints, while addressing the issue of global health, human ingenuity has proven again that in times of crisis, it is our best asset. Constraints like limited testing capacity and lack of real-time information regarding the spread of the virus, are the highest priority in the mitigation process, aside from the development of vaccines and the pushing through of vaccination programs. Using the available Chest X-Ray Images dataset and an AI-Computer Vision Technique called Convolutional Neural Network, features of the images were extracted and classified as COVID-19 positive or not. This paper proposes the usage of the 18-layer Residual Neural Network (ResNet-18) as an architecture instead of other ResNet with a higher number of layers. The researcher achieves the highest validation accuracy of 99.26%. Moving forward, using this lower number of layers in training a model classifier, resolves the issue of device constraints such as storage capacity and computing resources while still assuring highly accurate outputs. © 2022 IEEE.

An Improved COVID-19 Lung X-Ray Image Classification Algorithm Based on ConvNeXt Network

Aiming at the new coronavirus that appeared in 2019, which has caused a large number of infected patients worldwide due to its high contagiousness, in order to detect the source of infection in time and cut off the chain of transmission, we developed a new Chest X-ray (CXR) image classification algorithm with high accuracy, simple operation and fast processing for COVID-19. The algorithm is based on ConvNeXt pure convolutional neural network, we adjusted the network structure and loss function, added some new Data Augmentation methods and introduced attention mechanism. Compared with other classical convolutional neural network classification algorithms such as AlexNet, ResNet-34, ResNet-50, ResNet-101, ConvNeXt-tiny, ConvNeXt-small and ConvNeXt-base, the improved algorithm has better performance on COVID dataset.

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.

Assessing an AI-based Smart Imagery Framing and Truthing (SIFT) system to assist radiologists annotating lung abnormalities on chest X-ray images for development of deep learning models

To assess a Smart Imagery Framing and Truthing (SIFT) system in automatically labeling and annotating chest X-ray (CXR) images with multiple diseases as an assist to radiologists on multi-disease CXRs. SIFT system was developed by integrating a convolutional neural network based-augmented MaskR-CNN and a multi-layer perceptron neural network. It is trained with images containing 307,415 ROIs representing 69 different abnormalities and 67,071 normal CXRs. SIFT automatically labels ROIs with a specific type of abnormality, annotates fine-grained boundary, gives confidence score, and recommends other possible types of abnormality. An independent set of 178 CXRs containing 272 ROIs depicting five different abnormalities including pulmonary tuberculosis, pulmonary nodule, pneumonia, COVID-19, and fibrogenesis was used to evaluate radiologists' performance based on three radiologists in a double-blinded study. The radiologist first manually annotated each ROI without SIFT. Two weeks later, the radiologist annotated the same ROIs with SIFT aid to generate final results. Evaluation of consistency, efficiency and accuracy for radiologists with and without SIFT was conducted. After using SIFT, radiologists accept 93% SIFT annotated area, and variation across annotated area reduce by 28.23%. Inter-observer variation improves by 25.27% on averaged IOU. The consensus true positive rate increases by 5.00% (p=0.16), and false positive rate decreases by 27.70% (p<0.001). The radiologist's time to annotate these cases decreases by 42.30%. Performance in labelling abnormalities statistically remains the same. Independent observer study showed that SIFT is a promising step toward improving the consistency and efficiency of annotation, which is important for improving clinical X-ray diagnostic and monitoring efficiency. © 2023 SPIE.

Optimal Synergic Deep Learning for COVID-19 Classification Using Chest X-Ray Images

A chest radiology scan can significantly aid the early diagnosis and management of COVID-19 since the virus attacks the lungs. Chest X-ray (CXR) gained much interest after the COVID-19 outbreak thanks to its rapid imaging time, widespread availability, low cost, and portability. In radiological investigations, computer-aided diagnostic tools are implemented to reduce intra-and inter-observer variability. Using lately industrialized Artificial Intelligence (AI) algorithms and radiological techniques to diagnose and classify disease is advantageous. The current study develops an automatic identification and classification model for CXR pictures using Gaussian Fil-tering based Optimized Synergic Deep Learning using Remora Optimization Algorithm (GF-OSDL-ROA). This method is inclusive of preprocessing and classification based on optimization. The data is preprocessed using Gaussian filtering (GF) to remove any extraneous noise from the image's edges. Then, the OSDL model is applied to classify the CXRs under different severity levels based on CXR data. The learning rate of OSDL is optimized with the help of ROA for COVID-19 diagnosis showing the novelty of the work. OSDL model, applied in this study, was validated using the COVID-19 dataset. The experiments were conducted upon the proposed OSDL model, which achieved a classification accuracy of 99.83%, while the current Convolutional Neural Network achieved less classification accuracy, i.e., 98.14%.

Performance Analysis of VGG-16 Deep Learning Model for COVID-19 Detection using Chest X-Ray Images

COVID-19 is an infectious disease caused by newly discovered coronavirus. Currently, RT-PCR and Rapid Testing are used to test a person against COVID-19. These methods do not produce immediate results. Hence, we propose a solution to detect COVID-19 from chest X-ray images for immediate results. The solution is developed using a convolutional neural network architecture (VGG-16) model to extract features by transfer learning and a classification model to classify an input chest X-ray image as COVID-19 positive or negative. We introduced various parameters and computed the impact on the performance of the model to identify the parameters with high impact on the model's performance. The proposed solution is observed to provide best results compared to the existing ones. © 2023 Bharati Vidyapeeth, New Delhi.

A narrative review on radiation risk from imaging for COVID-19: Breaking the myths and the mithya

Radiotherapy-induced secondary malignancy is a well-known occurrence. During the COVID-19 pandemic, many people have undergone serial computed tomography (CT) imaging, and concerns have been raised regarding radiation-induced malignancies due to frequent scanning. Accordingly, various low and ultra-low-dose CT (LDCT) thorax protocols have been developed to reduce the dose of radiation. Major governing bodies worldwide have established guidelines regarding the indications for CT scans and chest X-rays during the pandemic. We, therefore, aimed to provide facts about the effects of radiation (both diagnostic and therapeutic). Through this article, we intend to break the myths and 'mithya' (misbeliefs) regarding diagnostic radiation and its association with cancer in this COVID-19 era. For this review, we performed a search in Google using specific keywords pertaining to imaging during COVID-19 and radiation risk. We also included the names of various global governing bodies in the Google search. We included only full text articles and guidelines from authentic websites. From this review, we conclude that if we follow the recommendations of various global governing bodies and use CT scan only in cases of moderate to severe COVID-related symptoms, adhere to the principle of 'as low as reasonably achievable' for radiation protection, and use LDCT scan protocols, we can significantly reduce the mean effective radiation dose delivered and the estimated cancer risk.Copyright © 2023 Cancer Research, Statistics, and Treatment. All rights reserved.

Decision region analysis to deconstruct the subgroup influence on AI/ML predictions

Assessing the generalizability of deep learning algorithms based on the size and diversity of the training data is not trivial. This study uses the mapping of samples in the image data space to the decision regions in the prediction space to understand how different subgroups in the data impact the neural network learning process and affect model generalizability. Using vicinal distribution-based linear interpolation, a plane of the decision region space spanned by the random 'triplet' of three images can be constructed. Analyzing these decision regions for many random triplets can provide insight into the relationships between distinct subgroups. In this study, a contrastive self-supervised approach is used to develop a 'base' classification model trained on a large chest x-ray (CXR) dataset. The base model is fine-tuned on COVID-19 CXR data to predict image acquisition technology (computed radiography (CR) or digital radiography (DX) and patient sex (male (M) or female (F)). Decision region analysis shows that the model's image acquisition technology decision space is dominated by CR, regardless of the acquisition technology for the base images. Similarly, the Female class dominates the decision space. This study shows that decision region analysis has the potential to provide insights into subgroup diversity, sources of imbalances in the data, and model generalizability. © 2023 SPIE.

Deep Convolutional Neural Network Based Automatic COVID-19 Detection from Chest X-ray Images

To minimize the rate of death from COVID-19 and stop the disease from spreading early detection is vital. The normal RT-PCR tests for COVID-19 detection take a long time to complete. In contrast to this test, Covid-19 can be quickly detected using various machine-learning technologies. Previous studies only had access to smaller datasets, as COVID-19 data was not readily available back then. Since COVID-19 is a dangerous virus, the model needs to be robust and trustworthy, and the model must be trained on a large and diverse dataset. To overcome that problem, this study combines six publicly available Chest X-ray datasets to produce a larger and more diverse balanced dataset with a total of 68,424 images. In this study, we develop a CNN model that primarily entails two steps: (a) feature extraction and (b) classification, which are used to identify COVID-19 positive cases from X-ray images. The accuracy of this proposed model is 97.58%, which is higher than most state-of-the-art models. © 2022 IEEE.

Rapid COVID-19 diagnosis using ensemble deep transfer learning models from chest radiographic images.

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) causes novel coronavirus disease (COVID-19) outbreak in more than 200 countries around the world. The early diagnosis of infected patients is needed to discontinue this outbreak. The diagnosis of coronavirus infection from radiography images is the fastest method. In this paper, two different ensemble deep transfer learning models have been designed for COVID-19 diagnosis utilizing the chest X-rays. Both models have utilized pre-trained models for better performance. They are able to differentiate COVID-19, viral pneumonia, and bacterial pneumonia. Both models have been developed to improve the generalization capability of the classifier for binary and multi-class problems. The proposed models have been tested on two well-known datasets. Experimental results reveal that the proposed framework outperforms the existing techniques in terms of sensitivity, specificity, and accuracy.

A comprehensive review of analyzing the chest X-ray images to detect COVID-19 infections using deep learning techniques.

COVID-19, a highly infectious respiratory disease a used by SARS virus, has killed millions of people across many countries. To enhance quick and accurate diagnosis of COVID-19, chest X-ray (CXR) imaging methods were commonly utilized. Identifying the infection manually by radio imaging, on the other hand, was considered, extremely difficult due to the time commitment and significant risk of human error. Emerging artificial intelligence (AI) techniques promised exploration in the development of precise and as well as automated COVID-19 detection tools. Convolution neural networks (CNN), a well performing deep learning strategy tends to gain substantial favors among AI approaches for COVID-19 classification. The preprints and published studies to diagnose COVID-19 with CXR pictures using CNN and other deep learning methodologies are reviewed and critically assessed in this research. This study focused on the methodology, algorithms, and preprocessing techniques used in various deep learning architectures, as well as datasets and performance studies of several deep learning architectures used in prediction and diagnosis. Our research concludes with a list of future research directions in COVID-19 imaging categorization.