A radiomics-based model for the outcome prediction in COVID-19 positive patients through deep learning with both longitudinal chest x-ray and chest computed tomography images

The coronavirus disease 2019 (COVID-19) pandemic had a major impact on global health and was associated with millions of deaths worldwide. During the pandemic, imaging characteristics of chest X-ray (CXR) and chest computed tomography (CT) played an important role in the screening, diagnosis and monitoring the disease progression. Various studies suggested that quantitative image analysis methods including artificial intelligence and radiomics can greatly boost the value of imaging in the management of COVID-19. However, few studies have explored the use of longitudinal multi-modal medical images with varying visit intervals for outcome prediction in COVID-19 patients. This study aims to explore the potential of longitudinal multimodal radiomics in predicting the outcome of COVID-19 patients by integrating both CXR and CT images with variable visit intervals through deep learning. 2274 patients who underwent CXR and/or CT scans during disease progression were selected for this study. Of these, 946 patients were treated at the University of Pennsylvania Health System (UPHS) and the remaining 1328 patients were acquired at Stony Brook University (SBU) and curated by the Medical Imaging and Data Resource Center (MIDRC). 532 radiomic features were extracted with the Cancer Imaging Phenomics Toolkit (CaPTk) from the lung regions in CXR and CT images at all visits. We employed two commonly used deep learning algorithms to analyze the longitudinal multimodal features, and evaluated the prediction results based on the area under the receiver operating characteristic curve (AUC). Our models achieved testing AUC scores of 0.816 and 0.836, respectively, for the prediction of mortality. © 2023 SPIE.

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.

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.

Robotic Assistance in Radiology: A Covid-19 Scenario

During the COVID-19 Pandemic, the need for rapid and reliable alternative COVID-19 screening methods have motivated the development of learning networks to screen COVID-19 patients based on chest radiography obtained from Chest X-ray (CXR) and Computed Tomography (CT) imaging. Although the effectiveness of developed models have been documented, their adoption in assisting radiologists suffers mainly due to the failure to implement or present any applicable framework. Therefore in this paper, a robotic framework is proposed to aid radiologists in COVID-19 patient screening. Specifically, Transfer learning is employed to first develop two well-known learning networks (GoogleNet and SqueezeNet) to classify positive and negative COVID-19 patients based on chest radiography obtained from Chest X-Ray (CXR) and CT imaging collected from three publicly available repositories. A test accuracy of 90.90%, sensitivity and specificity of 94.70% and 87.20% were obtained respectively for SqueezeNet and a test accuracy of 96.40%, sensitivity and specificity of 95.50% and 97.40% were obtained respectively for GoogleNet. Consequently, to demonstrate the clinical usability of the model, it is deployed on the Softbank NAO-V6 humanoid robot which is a social robot to serve as an assistive platform for radiologists. The strategy is an end-to-end explainable sorting of X-ray images, particularly for COVID-19 patients. Laboratory-based implementation of the overall framework demonstrates the effectiveness of the proposed platform in aiding radiologists in COVID-19 screening. Author

Deep Learning-and Transfer Learning-based Models for COVID-19 Detection using Radiography Images

Image classification and segmentation techniques are still very popular in the medical field (for healthcare), in which the medical image plays an important role in the detection and screening of diseases. Recently, the spread of new viral diseases, namely Covid-19, requires powerful computer models and rich resources (datasets) to fight this phenomenon. In this study, we propose to examine the CNN Deep Learning algorithm and two Transfer Learning models, namely RestNet50 and MobileNetV2 using the pretrained model of the ImageNet database, experimented on the new dataset (COVID-QU-Ex Dataset 2022) offered by the University of Qatar. These models are tested to classify radiography images into two classes (Covid19 and Normal). The results achieved by CNN (Acc =95.97%), ResNet50 (Acc =95.53%) and MobileNetV2 (Acc =97.32%) show that these algorithms are promising in order to combat this Covid-19 disease by detecting it through thoracic images (Chest X-ray type). © 2023 IEEE.

A DEEP LEARNING APPROACH LUNG SEGMENTATION AND PNEUMONIA DETECTION FROM X-RAYS

Lung segmentation is a process of detection and identification of lung cancer and pneumonia with the help of image processing techniques. Deep learning algorithms can be incorporated to build the computer-aided diagnosis (CAD) system for detecting or recognizing broad objects like acute respiratory distress syndrome (ARDS), Tuberculosis, Pneumonia, Lung cancer, Covid, and several other respiratory diseases. This paper presents pneumonia detection from lung segmentation using deep learning methods on chest radiography. Chest X-ray is the most useful technique among other existing techniques, due to its lesser cost. The main drawback of a chest x-ray is that it cannot detect all problems in the chest. Thus, implementing convolutional neural networks (CNN) to perform lung segmentation and to obtain correct results. The 'lost' regions of the lungs are reconstructed by an automatic segmentation method from raw images of chest X-ray. © 2022 IEEE.

A Deep Net Model for Covid-19 Detection Based on ROC Analysis Using Radiographs

The European Centre of Disease Prevention & Control's analytical statistics show that the new corona virus (Covid-19) is rapidly spreading amongst millions of people & causing the deaths of thousands of them. Despite the daily increase in cases, there are still a finite quantity of Covid-19 test kits available. The use of an automatic recognition system is crucial for the diagnosis and control of Covid-19. Three important Inception-ResNetV2, InceptionV3, & ResNet50 models of convolutional neural networks are utilized to detect the Corona Virus in lung X-ray radiography. The ResNet50 version has the best result & accuracy rate of the present system. As compared to the current models, a novel procedures and ensuring on the CNN model delivers better specific, sensitivities, and precision. By using confusion matrix and ROC assessment, fivefold validation data is utilized to analyze the current models and compare them to the proposed system. © 2023 IEEE.

A Lightweight CNN Architecture for Chest X-ray Images Analysis to Facilitate Early Stage Detection of Lung Diseases

Lung-related diseases are one of the significant causes of death among infants and children. However, the mortality rate can be reduced by the detection of lung abnormality at an early stage. Traditionally, radiologists identify irregularities by interpreting chest x-ray images which is time-consuming. Therefore, researchers have proposed many automated systems for diagnosing pneumonia and other lung-related diseases. Due to the remarkable performance of Convolutional Neural Networks(CNN) in image classification, it has gained immense popularity in chest x-ray image analysis. Most of the research has utilized famous pre-trained Imagenet models for more accurate analysis of Chest X-ray images. However, the problem with these architectures is that they have many parameters that increase the training time, which makes the detection process lengthy. This paper introduces a lightweight, compact, and well-tuned CNN architecture with far fewer parameters than the pre-trained model to analyze two of the most common lung diseases, pneumonia and Covid-19. We have evaluated our model on two benchmark datasets. Experimental results show that our lightweight CNN model has far fewer hyperparameters than other state-of-the-art models but achieves similar results. We have achieved an accuracy of 90.38% on the kermany dataset and 96.90% on the Covid-19 Radiography dataset. © 2022 IEEE.

A Transfer Learning Approach for Diagnosis of COVID-19 Cases from Chest Radiography Images

The new coronavirus disease 2019 (COVID-19) pandemic completely changed individuals' daily lives and created economic disruption across the world. Many countries are using movement restrictions and physical distancing as their measures to slow down this transmission. Effective screening of COVID-19 cases is needed to stop the spreading of these diseases. In the first phases of clinical assessment, it was seen that patients with deformities in chest X-ray images show the signs of COVID-19 infection. Inspired from this, in this study, a novel framework is designed to detect the COVID-19 cases from chest radiography images. Here, a pre-trained deep convolutional neural network VGG-16 is used to extract discriminating features from the radiography images. These extracted features are given as an input to the Logistic regression classifier for automatic detection of COVID-19 cases. The suggested framework obtained a remarkable accuracy of 99.1% with a 100% sensitivity rate in comparison with other state-of-the-art classifier. © 2022 IEEE.

Medical Diagnosis and Identification of Covid -19 by Intelligent IoT System and Resnet 18 Bilinear Deep Greedy Network

An international health crisis has been caused by the widespread COVID-19 epidemic. COVID-19 patient diagnoses are made using deep learning, although this necessitates a massive radiography data collection in order to efficiently deliver an optimum result. This paper presents a novel Intelligent System with IoT sensors for covid 19 and "Bilinear Resnet 18 Deep Greedy Network,"which is effective with a limited amount of datasets. Despite peculiarities brought on by a small dataset, the suggested approach could successfully combat the anomalies of over fitting and under fitting. The suggested architecture ensures a successful conclusion when the trained model is correctly evaluated using the provided X-ray datasets of COVID-19 cases. The recommended model offers accuracy of 97%, which is superior to existing methodologies. Better precision, recall, and F1 score are provided;which are 98%, 96%, and 96.94% respectively, which is better than other existing methodology. © 2022 IEEE.

COVID Detection Using Chest X-ray Images Using Ensembled Deep Learning

COVID-19 originated in Wuhan, China, in December 2019, and there have been over 464.5 million infected cases, and 6.08 million individuals have died worldwide. Effective detection of COVID-19 has been an essential task for stopping its quick spread and ultimately saving precious lives. This paper considers radiological examination using chest X-rays as patients with COVID-19 infections are likely to be adequately recognized using chest radiography pictures. Although many machine learning/deep learning techniques have been developed, their approach is likely to suffer problems like generalization error, high variance, overfitting, etc., due to limited dataset size. By producing predictions with numerous models rather than only one model, the ensemble model can overcome the disadvantages of deep learning. So, in this paper, we propose an ensemble deep learning method for detecting COVID-19 using chest X-ray images. On a combination of DenseNet, InceptionV3, and MobileNet, we got the best validation accuracy of 96.20% and testing accuracy of 92.45%. We hope this approach will help detect COVID-19 early and reduce further spread. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Classification of Chest Radiography Scans for COVID-19

Humanity has suffered as a result of the COVID-19 pandemic for more than two years. Testing kits were not widely accessible during the pandemic, which caused alarm. Any technical development that enables a quicker and more accurate identification of COVID-19 infection can be very beneficial for the medical field. X-rays can be used to examine a patient's lungs since COVID-19 targets the epithelial cells that line the respiratory system. It is challenging to determine COVID-19 from other Viral Pneumonia cases, though. The purpose of this paper is to examine the effectiveness of deep learning models in the quick and precise detection of COVID-19 in chest X-ray scans. © 2022 IEEE.

System for coronavirus classification based on a new textural descriptor

The combination of Chest X-Ray imaging and Artificial Intelligence (AI) has proven its efficiency in coronavirus disease (COVID-19) detection [1]. The present paper proposes an efficient COVID-19 detection system based on a new textural features descriptor: Monogenic Local Binary Pattern Variance (MLBPV). An Artificial Neural Network (ANN) model is used for Regions Of Interest (ROIs) classification. Evaluating MLBPV, it outperforms other tested models by achieving an Area Under Curve (A-{z}) of 0.96263 and an accuracy of 99.9805%. Comparing our method with previous ones proves that ours provides the best performance. This model may be implemented in digital X-Ray machine for radiography and help radiologists. © 2022 IEEE.

A Survey of Deep Learning Techniques for the Analysis of COVID-19 and their usability for Detecting Omicron

The Coronavirus (COVID-19) outbreak in December 2019 has drastically affected humans worldwide, creating a health crisis that has infected millions of lives and devastated the global economy. COVID-19 is ongoing, with the emergence of many new strains. Deep learning (DL) techniques have proven helpful in efficiently analysing and delineating infectious regions in radiological images. This survey paper draws a taxonomy of deep learning techniques for detecting COVID-19 infection in radiographic imaging modalities Chest X-Ray, and Computer Tomography. DL techniques are broadly categorised into classification, segmentation, and multi-stage approaches for COVID-19 diagnosis at the image and region-level analysis. These techniques are further classified as pre-trained and custom-made Convolutional Neural Network architectures. Furthermore, a discussion is drawn on radiographic datasets, evaluation metrics, and commercial platforms provided for detection. In the end, a brief look is paid to emerging ideas, gaps in existing research, and challenges in developing diagnostic techniques. This survey provides insight into the promising areas of research in DL and is likely to guide the research community on the upcoming development of deep learning techniques for COVID-19. This will pave the way to accelerate the research in designing customised DL-based diagnostic tools for effectively dealing with new variants of COVID-19 and emerging challenges. © 2023 Informa UK Limited, trading as Taylor & Francis Group.

Assessment of Chest X-Ray Image-Based COVID-19 Detection using Deep Transfer Learning Models

Wuhan is the city in China where COVID-19 was first discovered, and the disease quickly spread throughout the world, affecting over 215 million people. Vaccination has been tried to control the disease effects. Many data scientists contributed and analyzed the disease using chest X-Rays and Computed Tomography (CT) scans in order to control it. The data collected from Chest X-rays have been proven to be extremely effective for screening COVID-19 patients, particularly in terms of resolving overcapacity in emergency departments and urgent-care centers. Our proposed approach towards COVID-19 research contribution consists of four transfer learning models i.e., MobileNet, DenseNet201, InceptionNetV2 and NasNetMobile. Grayscale images of chest X-Rays that have been preprocessed are fed into these models as input data. The dataset used in the proposed framework is the COVID19 Radiography Database, which is available to all researchers on the Kaggle platform and contains four different types of chest X-ray images i.e., COVID-19, Pneumonia, Opacity and Normal. For multiclass classification that is MobileNet, DenseNet201, InceptionNetV3 and NasNetMobile the models showed an impressive accuracy of 91.26%, 90.38%, 89.27, and 87.74, while for binary class classification, the prediction capability of our used models is 97.03%, 96.78%, 95.18% and 95.40% respectively. © 2022 IEEE.

Covid-19 classification using CNN with radiography image data set

COVID-19 is a novel coronavirus disease that has been reported in Wuhan, China since late December 2019 and has subsequently spread around the world. In severe cases of illness, there may be no option but to die due to substantial alveolar damage and progressive respiratory failure. Testing with RT-PCR, for instance, is the gold standard for clinical diagnosis, but it is possible for the tests to produce false negatives. Further, the lack of resources for conducting RT-PCR testing may deter the next clinical decision and treatment under the pandemic situation. As a result, chest CT imaging has become a valuable tool for diagnostic and prognostic purposes in COVID-19 patients. Detection of COVID-19 early enables the development of prevention plans and a disease control plan. Through this experimentation, the main objective is to utilize transfer learning to leverage pre-trained weights from CNNs. We propose the ResNet50 architecture based on the ImageNet pre-trained weights to detect the Covid-19. The proposed model is evaluated on X-ray images of COVID-19 chests and on images taken with a Computerized Tomography scanner. Using the 746 images of covid and non-covid patient datasets are bifurcated into train and test datasets for training and validate our model and achieved 84.90 % model accuracy. The Accuracy, precision, recall and F1-Scores are presented along with the receiver operating characteristic (ROC) curve, the precision-recall curve, the average prediction, and the confusion matrix of three distinct models. © 2022 IEEE.

Convolutional Neural Network based COVID-19 Recognition from X-Ray Image

To halt extreme spread of Coronavirus(COVID-19), proper detection is the need of the hour. The number of physicians is negligible to serve the immense number of COVID-19 affected patients. For this reason, it is essential to automate the detection system of COVID-19 disease. In this proposed work, a convolutional neural network (CNN) based COVID-19 diagnosis system is developed to automate COVID- 19 disease detection using images of chest X-rays. The proposed model can differentiate three varieties: COVID-19, pneumonia and normal(healthy) from the X-ray images. Experimental process has been performed upon two publicly available datasets: COVID-19 Radiography Database and COVID-5K. Five deep convolutional neural network architectures: Xception, ResNet-50, Inception-v1, Inception-v2, and Inception-v3 are discretely used to train the system. The evaluation of the proposed system proves that Xception has provided the best performance with 99.47% accuracy, 99.21% sensitivity, 99.60% specificity, and 99.21% F1- score. The resultant of the experiment illustrates an improvement in the performance compared to some existing research works. © 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.

Comparison of Transfer Learning Models for COVID-19 Detection using Chest X-Ray Images

The transmission of cases of the COVID-19 virus (Coronavirus disease) is currently still racing. Even though the COVID-19 recovery rate has increased, new variants such as Omicron or Centaurus are still spreading in various countries. Detection COVID-19 based on Chest X-Ray is needed to avoid the wider transmission of the virus. This study uses the CNN (Convolutional Neural Network) transfer learning models like AlexNet, VGG19, Resnet50, InceptionV3 and the BasicCNN model. This study uses a dataset named the COVID-19 Radiography Database which contains three classes, namely COVID-19, Normal, and Viral Pneumonia. The results of this study indicate that the Resnet50 model is the best model that produces the highest accuracy compared to other models. The Resnet50 model obtained an accuracy of 98.68%. Then followed by other models in sequence, namely InceptionV3, VGG19, AlexNet, and BasicCNN. Evaluation in this study also uses Precision, Recall, and F1-Measure which show that Resnet50 obtains the highest value compared with other methods. This shows that the transfer learning model has a good performance for detection of COVID-19 based on Chest X-Ray. © 2022 IEEE.

COVID-19 Classification from X-Ray Images using 2D CNN

The coronavirus (COVID-19) detection has been a crucial task for researchers, scientists, health experts all across the world and everyone is trying together to find a solution to it. The X-rays images of lungs have become one of the most prevalent and effective procedures used by researchers to monitor COVID-19. Unfortunately, inspecting each case involves multiple radiology experts and time, which is one of the critical tasks in such an outbreak. In this paper, a deep learning approach, 2D convolutional neural networks (CNN) has been used to classify healthy and COVID-19 chest X-ray images. 'Curated Dataset for COVID-19 Posterior-Anterior Chest Radiography Images (X-Rays)' dataset has been used in this study. The major indicator of this study is the accuracy of the proposed model. The classification model, 2D CNN has achieved accuracy and f1-score of 0.96 and 0.95 respectively. © 2022 IEEE.