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.

Assessing Robustness of a Deep Learning Model for COVID-19 Classification on Chest Radiographs

The COVID-19 pandemic has made a dramatic impact on human life, medical systems, and financial resources. Due to the disease's pervasive nature, many different and interdisciplinary fields of research pivoted to study the disease. For example, deep learning (DL) techniques were employed early to assess patient diagnosis and prognosis from chest radiographs (CXRs) and computed tomography (CT) scans. While the use of artificial intelligence (AI) in the medical sector has displayed promising results, DL may suffer from lack of reproducibility and generalizability. In this study, the robustness of a pre-trained DL model utilizing the DenseNet-121 architecture was evaluated by using a larger collection of CXRs from the same institution that provided the original model with its test and training datasets. The current test set contained a larger span of dates, incorporated different strains of the virus, and included different immunization statuses. Considering differences in these factors, model performance between the original and current test sets was evaluated using area under the receiver operating characteristic curve (ROC AUC) [95% CI]. Statistical comparisons were performed using the Delong, Kolmogorov-Smirnov, and Wilcoxon rank-sum tests. Uniform manifold approximation and projection (UMAP) was used to help visualize whether underlying causes were responsible for differences in performance between test sets. In the task of classifying between COVID-positive and COVID-negative patients, the DL model achieved an AUC of 0.67 [0.65, 0.70], compared with the original performance of 0.76 [0.73, 0.79]. The results of this study suggest that underlying biases or overfitting may hinder performance when generalizing the model. © 2023 SPIE.

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.

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.

Identification and Localization of COVID-19Abnormalities on Chest Radiographs using Computer Vision

The Corona Virus (COVID 19) pandemic is quickly becoming the world's most deadly disease. The spreading rate is higher and the early detection helps in faster recovery. The existence of COVID 19 in individuals shall be detected using molecular analysis or through radiographs of lungs. As time and test kit are limited RT- PCR is not suitable to test all. The RT- PCR being a time-consuming process, diagnosis using chest radiographs needs no transportation as the modern X-ray systems are digitized. Deep learning takes an edge over other techniques as it deduces the features automatically and performs massively parallel computations. Multiple feature maps will help in accurate prediction. The objective of the proposed work is to develop a Computer Aided Deep Learning System identify and localize COVID-19 virus from other viruses and pneumonia. It helps to detect COVID-19 within a short period of time thereby improving the lifetime of the individuals. SIIM-FISABIO-RSNA benchmark datasets are used to examine the proposed system. Recall, Precision, Accuracy-rate, and F-Measure are the metrics used to prove the integrity of the system. © 2023 IEEE.

A Transfer Learning Method for Covid-19 and Pneumonia Diagnosis Based on Chest Radiograph Classification

Pneumonia has been a tough and dangerous human illness for a history-long time, notably since the COVID-19 pandemic outbreak. Many pathogens, including bacteria or viruses like COVID-19, can cause pneumonia, leading to inflammation in patients' alveoli. A corresponding symptom is the appearance of lung opacities, which are vague white clouds in the lungs' darkness in chest radiographs. Modern medicine has indicated that pneumonia-associated opacities are distinguishable and can be seen as fine-grained labels, which make it possible to use deep learning to classify chest radiographs as a supplementary aid for disease diagnosis and performing pre-screening. However, deep learning-based medical imaging solutions, including convolutional neural networks, often encounter a performance bottleneck when encountering a new disease due to the dataset's limited size or class imbalance. This study proposes a deep learning-based approach using transfer learning and weighted loss to overcome this problem. The contributions of it are three-fold. First, we propose an image classification model based on pre-trained Densely Connected Convolutional Networks using Weighted Cross Entropy. Second, we test the effect of masking non-lung regions on the classification performance of chest radiographs. Finally, we summarize a generic practical paradigm for medical image classification based on transfer learning. Using our method, we demonstrate that pre-training on the COVID-19 dataset effectively improves the model's performance on the non-COVID Pneumonia dataset. Overall, the proposed model achieves excellent performance with 95.75% testing accuracy on a multiclass classification for the COVID-19 dataset and 98.29% on a binary classification for the Pneumonia dataset. © 2023 IEEE.

COVIDEffiNet: Pulmonary Diseases and COVID-19 Detection from Chest Radiographs Using EfficientNet Deep Learning Model

There have been attempts made previously to classify and determine the diagnosis of a disease of a patient based on the X-rays and computed tomography images of various parts of the body. In the field of lung disease diagnosis, there have been attempts to identify lungs infected with pneumonia, COVID-19, and tuberculosis, either individually classifying them into two groups of positive and negative of the given disease or in groups with multiple classes. These methods and approaches have used various deep learning models like CNNs, ResNet50, VGG19, Inception V3, MobileNet_V2, hybrid models, and ensemble learning methods. In this paper, we have proposed a model that takes an X-ray image of the lungs of the patients as input and classifies the result as one of the following classes: tuberculosis, pneumonia, COVID-19, or normal, that is, healthy lungs. What we have used here is transfer learning, with our base model being EfficientNet which gives an accuracy of 93%. For this, we have used different datasets of X-ray images of patients with different lung ailments, namely pneumonia, tuberculosis, and COVID. The dataset consists of images in four categories, the above-mentioned three diseases and a fourth category of normal healthy lungs. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

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.

Artificial intelligence: A key for detecting COVID-19 using chest radiography

An impending branch of computer science is artificial intelligence. It plays an important role in the construction of smart machines that are capable of performing sophisticated operations. One of the key characteristics of artificial intelligence is its ability to make decisions on its own and rationalize the solution, helping us to achieve a certain goal. Our human race has faced many threats in the form of epidemics and pandemics, which have proved to be almost incurable in the past. Nevertheless, science and its evolving technologies have given us some hope to fight such threats. One such pandemic that our human race is facing in the current times is COVID-19. This deadly disease is rapidly spreading across the whole world endangering the lives of humans. Amid the chaos, we desperately need to stop the spread, or at least take adequate counter-active measures to detect this virus at its early stage. Deep learning, a subset of artificial intelligence provides many models which helps in the automation of the task of detecting viruses in humans mainly with the help of image processing. In detecting COVID-19, deep learning is a breakthrough, which has helped us in our proposed system. This system makes use of chest radiographs (CXR) to detect the presence of the virus in the human body thereby lowering the risk of spread which is fairly high in manual detection methods. The CXRs are one of the most common imaging tests in the clinical field, which helps in detecting the presence of cold, cough, shortness of breath in the lungs, and so on. The proposed model is very efficient when it comes to detecting problems in the lungs with the help of image processing. We propose an improvised neural network derived from the Convolutional Neural Network which works similar to the human brain structure to detect and process the CXR images efficiently and at faster rates. The neural network mimics the functioning of the brain, where self-learning and decision making are its key features. The image data sets are a collection of CXR images which have a RGB value of 1. This approach is proven to be safer and better than the manual testing methods that are currently deployed. As the traditional methods for detecting COVID-19 virus is tedious, and not fairly accurate, automating this task can help in giving accurate results with reduced risk of spread of disease through physical contact. © 2022 Scrivener Publishing LLC.

Firefly Optimization with Bidirectional Gated Recurrent Unit for COVID-19 Diagnosis on Chest Radiographs

The epidemic of coronavirus disease 2019 (COVID-19) has caused an ever-growing demand for treatment, testing, and diagnosis. Chest x-rays are a fast and low-cost test that can detect COVID19 but chest imaging is not a first-line test for COVID19 because of lower diagnosis performance and confounding with other viral pneumonia. Current studies using deep learning (DL) might assist in overcoming these issues as convolution neural networks (CNN) have illustrated higher performance of COVID19 diagnoses at the earlier phase. This study develops a new Firefly Optimization with Bidirectional Gated Recurrent Unit (FFO-BGRU) for COVID19 diagnoses on Chest Radiographs. The main intention of the FFO-BGRU technique lies in the recognition and classification of COVID-19 on Chest X-ray images. At the initial stage, the presented FFO-BGRU technique applies Wiener filtering (WF) technique for noise removal process. Followed, the hyperparameter tuning process takes place by using FFO algorithm and SqueezeNet architecture is applied for feature extraction. Lastly, the BGRU model is applied for COVID19 recognition and classification. A wide range of simulations were performed to demonstrate the betterment of the FFO-BGRU model. The comprehensive comparison study highlighted the improved outcomes of the FFO-BGRU algorithm over other recent approaches. © 2022 IEEE.

COVID-19 radiograph prognosis using a deep CResNeXt network.

COVID-19 has caused an epidemic in the entire world and it is caused by the novel virus SARS-COV-2. In severe conditions, this virus can cause a critical lung infection or viral pneumonia. To administer the correct treatment to patients, COVID-19 testing is important for diagnosing and determining patients who are infected with COVID-19, as opposed to those infected with other bacterial or viral infections. In this paper, a CResNeXt chest radiograph COVID-19 prediction model is proposed using residual network architecture. The advantage of the proposed model is that it requires lesser free hyper-parameters as compared to other residual networks. In addition, the training time per epochs of the model is very less compared to VGG19, ResNet-50, ResNeXt. The proposed CResNeXt model's binary classification (COVID-19 versus No-Finding) accuracy is observed to be 98.63% and 99.99% and multi-class classification (COVID-19, Pneumonia, and No-Finding) accuracy is observed to be 97.42% and 99.27% on the original and augmented datasets, respectively.

CXR Segmentation by AdaIN-Based Domain Adaptation and Knowledge Distillation

As segmentation labels are scarce, extensive researches have been conducted to train segmentation networks with domain adaptation, semi-supervised or self-supervised learning techniques to utilize abundant unlabeled dataset. However, these approaches appear different from each other, so it is not clear how these approaches can be combined for better performance. Inspired by recent multi-domain image translation approaches, here we propose a novel segmentation framework using adaptive instance normalization (AdaIN), so that a single generator is trained to perform both domain adaptation and semi-supervised segmentation tasks via knowledge distillation by simply changing task-specific AdaIN codes. Specifically, our framework is designed to deal with difficult situations in chest X-ray radiograph (CXR) segmentation, where labels are only available for normal data, but the trained model should be applied to both normal and abnormal data. The proposed network demonstrates great generalizability under domain shift and achieves the state-of-the-art performance for abnormal CXR segmentation.

Radiological Diagnosis of SARS-CoV-2 Infected Patients by Automated Classification of Chest Radiographs using DTL

SARS-CoV-2 is a strain of coronavirus in the Orthocoronavirinae subfamily, which was first identified in Wuhan, China, in December 2019. COVID-19 is the name of the disease produced by SARS-CoV-2. Because SARS-CoV-2 is cytopathic to airway epithelial cells and alveolar cells, researchers showed that combining features extracted from radiographic images with laboratory results may significantly aid in the early detection of the infection. Standard features seen in radiographs of patients with coronavirus are infiltrated or patchy opacities, like several features of other forms of viral pneumonia. In the early stages of COVID-19 infection, no abnormalities can be observed on radiographic images. However, as the disease unfolds, COVID-19 progressively manifests as a typical unilateral inflammation involving the middle and upper or lower lungs, sometimes with consolidation. Almost every hospital has at least one radiographer routinely used for diagnosing pneumonia, lymph nodes, and other conditions. Unfortunately, acquiring radiographic images and analyzing COVID-19 analyzing by a radiologist is time-consuming. Therefore, this work aims to assess the detection capability of the viral respiratory disease caused by the SARSCoV-2 virus from radiographic images using a deep transfer learning model implemented in a development environment for numerical computation and statistical analysis called MATLAB. © 2022 IEEE.

Autoencoder-based Unsupervised Anomaly Detection for Covid-19 Screening on Chest X-Ray Images

The COVID-19 outbreak is a major global catastrophe of our time and the largest hurdle since World War II. According to WHO, as of July 2022, there are more than 571 million confirmed cases of COVID-19 and over six million deaths. The issue of identifying unexpected inputs based on trained examples of normal data is known as anomaly detection. In the case of diagnosing covid-19, Chest X-ray disorders that are hardly apparent are extremely challenging to identify. Although various well-known supervised classification methods are being applied for that purpose, however in the real scenario, healthy patients' data is tremendously available but contaminated samples are scarce. The process of gathering samples from ill patients is troublesome and takes a lengthy time. To address the issue of data imbalance in anomaly detection, this research demonstrates an unsupervised learning technique using a convolutional autoencoder in which the training phase does not include any infected sample. Being trained only with the healthy data, The patterns of the healthy samples are preserved in latent vector space and can differentiate ill samples by observing substantial divergence from the distribution of healthy data. Higher reconstruction error and lower KDE (Kernel Density Estimation) indicate affected data. By contrasting the reconstruction error and KDE of healthy data with anomalous data, the suggested technique is feasible for identifying anomalous samples. © 2022 IEEE.

A Holistic Approach to Identify and Classify COVID-19 from Chest Radiographs, ECG, and CT-Scan Images Using ShuffleNet Convolutional Neural Network.

Early and precise COVID-19 identification and analysis are pivotal in reducing the spread of COVID-19. Medical imaging techniques, such as chest X-ray or chest radiographs, computed tomography (CT) scan, and electrocardiogram (ECG) trace images are the most widely known for early discovery and analysis of the coronavirus disease (COVID-19). Deep learning (DL) frameworks for identifying COVID-19 positive patients in the literature are limited to one data format, either ECG or chest radiograph images. Moreover, using several data types to recover abnormal patterns caused by COVID-19 could potentially provide more information and restrict the spread of the virus. This study presents an effective COVID-19 detection and classification approach using the Shufflenet CNN by employing three types of images, i.e., chest radiograph, CT-scan, and ECG-trace images. For this purpose, we performed extensive classification experiments with the proposed approach using each type of image. With the chest radiograph dataset, we performed three classification experiments at different levels of granularity, i.e., binary, three-class, and four-class classifications. In addition, we performed a binary classification experiment with the proposed approach by classifying CT-scan images into COVID-positive and normal. Finally, utilizing the ECG-trace images, we conducted three experiments at different levels of granularity, i.e., binary, three-class, and five-class classifications. We evaluated the proposed approach with the baseline COVID-19 Radiography Database, SARS-CoV-2 CT-scan, and ECG images dataset of cardiac and COVID-19 patients. The average accuracy of 99.98% for COVID-19 detection in the three-class classification scheme using chest radiographs, optimal accuracy of 100% for COVID-19 detection using CT scans, and average accuracy of 99.37% for five-class classification scheme using ECG trace images have proved the efficacy of our proposed method over the contemporary methods. The optimal accuracy of 100% for COVID-19 detection using CT scans and the accuracy gain of 1.54% (in the case of five-class classification using ECG trace images) from the previous approach, which utilized ECG images for the first time, has a major contribution to improving the COVID-19 prediction rate in early stages. Experimental findings demonstrate that the proposed framework outperforms contemporary models. For example, the proposed approach outperforms state-of-the-art DL approaches, such as Squeezenet, Alexnet, and Darknet19, by achieving the accuracy of 99.98 (proposed method), 98.29, 98.50, and 99.67, respectively.

Augmentation of literature review of COVID-19 radiology.

We suggest an augmentation of the excellent comprehensive review article titled "Comprehensive literature review on the radiographic findings, imaging modalities, and the role of radiology in the coronavirus disease 2019 (COVID-19) pandemic" under the following categories: (1) "Inclusion of additional radiological features, related to pulmonary infarcts and to COVID-19 pneumonia"; (2) "Amplified discussion of cardiovascular COVID-19 manifestations and the role of cardiac magnetic resonance imaging in monitoring and prognosis"; (3) "Imaging findings related to fluorodeoxyglucose positron emission tomography, optical, thermal and other imaging modalities/devices, including 'intelligent edge' and other remote monitoring devices"; (4) "Artificial intelligence in COVID-19 imaging"; (5) "Additional annotations to the radiological images in the manuscript to illustrate the additional signs discussed"; and (6) "A minor correction to a passage on pulmonary destruction".

Identification and Localization of COVID-19 Abnormalities on Chest Radiographs using Ensembled Deep Neural Networks

With the hit of the global pandemic COVID-19, the chest X-ray domain has gained prominence. It has been recognised as one of the principal methods to learn the presence of infection and its effect on various internal organs like the lungs. Chest radiographs show abnormalities due to COVID-19 that appear similar to the anomalies caused by other viruses and bacteria, thus making it challenging for technicians to detect. Therefore, it becomes almost inevitable to have a computer vision model that identifies and localizes the COVID-19 virus to help doctors provide an immediate and confident diagnosis. The models in computer vision tasks have seen considerable advancements in deep learning, so the proposed model tried to integrate a few of them to come up with a model for classifying and localising the diagnosis of COVID-19 using chest X-rays. This paper ensembles a few state-of-the-art models in classification and object detection to build a model for chest radiograph diagnosis. The proposed ensembled model is found to achieve the mean Average Precision value of 0.627 on SIIM-FISABIO-RSNA COVID-19 dataset. © 2022 IEEE.

Convolution Filter-Based Deep Neural Networks for Timely Diagnosis of COVID-19 Disease with Chest Radiographs

COVID-19 is a deadly disease that spreads in the lungs and may result in damaging both the lungs. This infection may be life threatening if not detected in time. In this work, chest radiographs after preprocessing were provided to the various pre-trained CNN models to extract the essential features. Once these features were extracted, the scans were provided to several classifiers for the classification of images as COVID-19 infected or normal chest radiographs. The result shows an accuracy of 99.5% and was obtained by using the VGG19 model and logistic regression (LR) classifier. The obtained values verify that the work conducted in this paper will be very useful in the timely analysis of COVID-19 infection to provide in-time treatment to the patients. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Mortality Prediction Analysis among COVID-19 Inpatients Using Clinical Variables and Deep Learning Chest Radiography Imaging Features.

The emergence of the COVID-19 pandemic over a relatively brief interval illustrates the need for rapid data-driven approaches to facilitate clinical decision making. We examined a machine learning process to predict inpatient mortality among COVID-19 patients using clinical and chest radiographic data. Modeling was performed with a de-identified dataset of encounters prior to widespread vaccine availability. Non-imaging predictors included demographics, pre-admission clinical history, and past medical history variables. Imaging features were extracted from chest radiographs by applying a deep convolutional neural network with transfer learning. A multi-layer perceptron combining 64 deep learning features from chest radiographs with 98 patient clinical features was trained to predict mortality. The Local Interpretable Model-Agnostic Explanations (LIME) method was used to explain model predictions. Non-imaging data alone predicted mortality with an ROC-AUC of 0.87 ± 0.03 (mean ± SD), while the addition of imaging data improved prediction slightly (ROC-AUC: 0.91 ± 0.02). The application of LIME to the combined imaging and clinical model found HbA1c values to contribute the most to model prediction (17.1 ± 1.7%), while imaging contributed 8.8 ± 2.8%. Age, gender, and BMI contributed 8.7%, 8.2%, and 7.1%, respectively. Our findings demonstrate a viable explainable AI approach to quantify the contributions of imaging and clinical data to COVID mortality predictions.

COVID-19 Detection from Chest Radiograph Images of Patients using Neural Networks

The fast escalation of COVID-19 since its outbreak has affected the health care system globally. To pare the dissemination of the disease it is important to detect COVID-19 positive cases at an early stage and isolate the patients as early as possible. So, an effective and timely method for detecting the disease is required. Despite the fact that RT-PCR is the benchmark for COVID-19 detection and is very accurate, it is time-consuming. Applying deep learning along with radiography may help in diagnosing patients with COVID-19 much fast. In the paper, we propose a deep learning-based CNN architecture inspired by AlexNet which takes a three-channel image input. © 2022 IEEE.