A Deep Ensemble Model for Differential Diagnosis of Covid-19 Pneumonia using Thoracic Radiography

The successful elimination of the SARS-Cov2 virus has evaded the society and medical fraternity to date. Months have passed but the virus is still very much present amongst us though its severity and contagiousness have decreased. The pandemic which was first detected in Wuhan, China in late 2019 has had colossal ramifications for the societal, financial and physical well-being of humankind. Timely detection and isolation of infected persons is the only way to contain this contagion. One of the biggest hurdles in accurately detecting Covid-19 is its similarities to other thoracic ailments such as Lung cancer, bacterial and viral Pneumonia, tuberculosis and others. Differential observation is challenging due to identical radioscopic discoveries such as GGOs, crazy paving structures and their combinations. Thorax imaging such as X-rays(CXR) have proven to be an efficient and economical diagnostics for detecting Covid-19 Pneumonia. The proposed work aims at utilising three CNN models namely Inception-V3, DenseNet169 and VGG16 along with feature concatenation and Ensemble technique to correctly predict Covid-19 Pneumonia from Chest X-rays of patients. The Covid-19 Radiography dataset, having a total of 4839 CXR images, has been employed to evaluate the proposed model and accuracy, precision, recall and F1-Score of 97.74%, 97.78%, 97.73% and 97.75% has been obtained. The proposed system can assist medical professionals in detecting Covid-19 from a host of other pulmonary diseases with a high probability. © 2023 IEEE.

Detecting COVID-19 from Chest X-rays Using Convolutional Neural Network Ensembles

Starting in late 2019, the coronavirus SARS-CoV-2 began spreading around the world and causing disruption in both daily life and healthcare systems. The disease is estimated to have caused more than 6 million deaths worldwide [WHO]. The pandemic and the global reaction to it severely affected the world economy, causing a significant increase in global inflation rates, unemployment, and the cost of energy commodities. To stop the spread of the virus and dampen its global effect, it is imperative to detect infected patients early on. Convolutional neural networks (CNNs) can effectively diagnose a patient's chest X-ray (CXR) to assess whether they have been infected. Previous medical image classification studies have shown exceptional accuracies, and the trained algorithms can be shared and deployed using a computer or a mobile device. CNN-based COVID-19 detection can be employed as a supplement to reverse transcription-polymerase chain reaction (RT-PCR). In this research work, 11 ensemble networks consisting of 6 CNN architectures and a classifier layer are evaluated on their ability to differentiate the CXRs of patients with COVID-19 from those of patients that have not been infected. The performance of ensemble models is then compared to the performance of individual CNN architectures. The best ensemble model COVID-19 detection accuracy was achieved using the logistic regression ensemble model, with an accuracy of 96.29%, which is 1.13% higher than the top-performing individual model. The highest F1-score was achieved by the standard vector classifier ensemble model, with a value of 88.6%, which was 2.06% better than the score achieved by the best-performing individual model. This work demonstrates that combining a set of top-performing COVID-19 detection models could lead to better results if the models are integrated together into an ensemble. The model can be deployed in overworked or remote health centers as an accurate and rapid supplement or back-up method for detecting COVID-19.

DeepCOVID-Fuse: A Multi-Modality Deep Learning Model Fusing Chest X-rays and Clinical Variables to Predict COVID-19 Risk Levels.

The COVID-19 pandemic has posed unprecedented challenges to global healthcare systems, highlighting the need for accurate and timely risk prediction models that can prioritize patient care and allocate resources effectively. This study presents DeepCOVID-Fuse, a deep learning fusion model that predicts risk levels in patients with confirmed COVID-19 by combining chest radiographs (CXRs) and clinical variables. The study collected initial CXRs, clinical variables, and outcomes (i.e., mortality, intubation, hospital length of stay, Intensive care units (ICU) admission) from February to April 2020, with risk levels determined by the outcomes. The fusion model was trained on 1657 patients (Age: 58.30 ± 17.74; Female: 807) and validated on 428 patients (56.41 ± 17.03; 190) from the local healthcare system and tested on 439 patients (56.51 ± 17.78; 205) from a different holdout hospital. The performance of well-trained fusion models on full or partial modalities was compared using DeLong and McNemar tests. Results show that DeepCOVID-Fuse significantly (p < 0.05) outperformed models trained only on CXRs or clinical variables, with an accuracy of 0.658 and an area under the receiver operating characteristic curve (AUC) of 0.842. The fusion model achieves good outcome predictions even when only one of the modalities is used in testing, demonstrating its ability to learn better feature representations across different modalities during training.

Classification of nucleic acid amplification on ISFET arrays using spectrogram-based neural networks.

The COVID-19 pandemic has highlighted a significant research gap in the field of molecular diagnostics. This has brought forth the need for AI-based edge solutions that can provide quick diagnostic results whilst maintaining data privacy, security and high standards of sensitivity and specificity. This paper presents a novel proof-of-concept method to detect nucleic acid amplification using ISFET sensors and deep learning. This enables the detection of DNA and RNA on a low-cost and portable lab-on-chip platform for identifying infectious diseases and cancer biomarkers. We show that by using spectrograms to transform the signal to the time-frequency domain, image processing techniques can be applied to achieve the reliable classification of the detected chemical signals. Transformation to spectrograms is beneficial as it makes the data compatible with 2D convolutional neural networks and helps gain significant performance improvement over neural networks trained on the time domain data. The trained network achieves an accuracy of 84% with a size of 30kB making it suitable for deployment on edge devices. This facilitates a new wave of intelligent lab-on-chip platforms that combine microfluidics, CMOS-based chemical sensing arrays and AI-based edge solutions for more intelligent and rapid molecular diagnostics.

CTMLP: Can MLPs replace CNNs or transformers for COVID-19 diagnosis?

BACKGROUND: Convolutional Neural Networks (CNNs) and the hybrid models of CNNs and Vision Transformers (VITs) are the recent mainstream methods for COVID-19 medical image diagnosis. However, pure CNNs lack global modeling ability, and the hybrid models of CNNs and VITs have problems such as large parameters and computational complexity. These models are difficult to be used effectively for medical diagnosis in just-in-time applications. METHODS: Therefore, a lightweight medical diagnosis network CTMLP based on convolutions and multi-layer perceptrons (MLPs) is proposed for the diagnosis of COVID-19. The previous self-supervised algorithms are based on CNNs and VITs, and the effectiveness of such algorithms for MLPs is not yet known. At the same time, due to the lack of ImageNet-scale datasets in the medical image domain for model pre-training. So, a pre-training scheme TL-DeCo based on transfer learning and self-supervised learning was constructed. In addition, TL-DeCo is too tedious and resource-consuming to build a new model each time. Therefore, a guided self-supervised pre-training scheme was constructed for the new lightweight model pre-training. RESULTS: The proposed CTMLP achieves an accuracy of 97.51%, an f1-score of 97.43%, and a recall of 98.91% without pre-training, even with only 48% of the number of ResNet50 parameters. Furthermore, the proposed guided self-supervised learning scheme can improve the baseline of simple self-supervised learning by 1%-1.27%. CONCLUSION: The final results show that the proposed CTMLP can replace CNNs or Transformers for a more efficient diagnosis of COVID-19. In addition, the additional pre-training framework was developed to make it more promising in clinical practice.

Investigation on MLP, CNNs and Vision Transformer models performance for Extracting a Human Emotions via Facial Expressions

One of the difficult challenges in AI development is to make machine understand the human feeling through expression because human can express feeling in various ways, for example, through voices, facial actions or behaviors. Facial Emotion Recognition (FER) has been used in interrogating suspects and being a tool to help detect emotions in people with nerve damage or even in the COVID-19 pandemic when patients hide their timelines. It can be applied to detect lies through micro expression. In this work will mainly focus on FER. The results of Deep Neural Network (DNN), Convolutional Neural Network (CNN), and Vision Transformer were compared. Human emotion expressions were classified by using facial expression datasets from AffectNet, Tsinghua, Extended Cohn Kanade (CK+), Karolinska Directed Emotional Faces (KDEF) and Real-world Affective Faces (RAF). Finally, all models were evaluated on the testing dataset to confirm their performance. The result shows that Vision Transformer model outperforms other models. © 2023 IEEE.

CNR-IEMN-CD and CNR-IEMN-CSD Approaches for Covid-19 Detection and Covid-19 Severity Detection from 3D CT-scans

Since its appearance in late 2019, Covid-19 has become an active research topic for the artificial intelligence (AI) community. One of the most interesting AI topics is Covid-19 analysis from medical imaging. CT-scan imaging is the most informative tool about this disease. This work is part of the 2nd COV19D competition, where two challenges are set: Covid-19 Detection and Covid-19 Severity Detection from the CT-scans. For Covid-19 detection from CT-scans, we proposed an ensemble of 2D Convolution blocks with Densenet-161 models (CNR-IEMN-CD). Here, each 2D convolutional block with Densenet-161 architecture is trained separately and in the testing phase, the ensemble model is based on the average of their probabilities. On the other hand, we proposed an ensemble of Convolutional Layers with Inception models for Covid-19 severity detection CNR-IEMN-CSD. In addition to the Convolutional Layers, three Inception variants were used, namely Inception-v3, Inception-v4 and Inception-Resnet. Our proposed approaches outperformed the baseline approach in the validation data of the 2nd COV19D competition by 11% and 16% for Covid-19 detection and Covid-19 severity detection, respectively. In the testing phase, our proposed approach CNR-IEMN-CD ranked fifth and improved the baseline results by 18.37%. On the other hand, our proposed approach CNR-IEMN-CSD ranked third in the test data of the 2nd COV19D competition for Covid-19 severity detection, and improved the baseline results by 6.81%. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

A deep learning approach for COVID-19 detection from computed tomography scans

The classification of COVID-19 patients from chest computed tomography (CT) images is a very difficult task due to the similarities observed with other lung diseases. Based on various CT scans of COVID and non-COVID patients, the aim of this chapter is to propose a simple deep learning architecture and compare its diagnostic performance using transfer learning and several machine learning techniques that could extract COVID-19's graphical features and classify them in order to provide a clinical diagnosis ahead of the pathogenic test, thus saving critical time for disease control. We also compare our approach and show that it outperforms various previous state-of-the-art techniques. We propose a deep learning architecture for transfer learning that is just a simple modification of eight new layers on the ImageNet pretrained convolutional neural networks (CNNs) which yielded us the best test accuracy of 98.30%, F1 score of 0.982, area under the receiver operating characteristic (ROC) curve of 0.982, and kappa value of 0.964 after training. Moreover, we use the proposed architecture for feature extraction and study the performance of various classifiers on them and were able to obtain the highest test accuracy of 91.75% with K-nearest neighbors. Also, we compare multiple CNNs and machine learning models for their diagnostic potential in disease detection and suggest a much faster and automated disease detection methodology. We show that smaller and memory efficient architectures are equally good compared to deep and heavy architectures at classifying chest CTs. We also show that visual geometry group (VGG) architectures are overall the best for this task. © 2023 Elsevier Inc. All rights reserved.

IoT-enabled stacked ensemble of deep neural networks for the diagnosis of COVID-19 using chest CT scans

The ongoing COVID-19 (novel coronavirus disease 2019) pandemic has triggered a global emergency, resulting in significant casualties and a negative effect on socioeconomic and healthcare systems around the world. Hence, automatic and fast screening of COVID-19 infections has become an urgent need of this pandemic. Real-time reverse transcription polymerase chain reaction (RT-PCR), a commonly used primary clinical method, is expensive and time-consuming for skilled health professionals. With the aid of various AI functionalities and advanced technologies, chest CT scans may thus be a viable alternative for quick and automatic screening of COVID-19. At the moment, significant advances in 5G cellular and internet of things (IoT) technology are finding use in various applications in the healthcare sector. This study presents an IoT-enabled deep learning-based stacking model to analyze chest CT scans for effective diagnosis of COVID-19 encounters. At first, patient data will be obtained using IoT devices and sent to a cloud server during the data procurement stage. Then we use different fine-tuned CNN sub-models, which are stacked together using a meta-learner to detect COVID-19 infection from input CT scans. The proposed model is evaluated using an open access dataset containing both COVID-19 infected and non-COVID CT images. Evaluation results show the efficacy of the proposed stacked model containing fine-tuned CNNs and a meta-learner in detecting coronavirus infections using CT scans. [ABSTRACT FROM AUTHOR] Copyright of Computing is the property of Springer Nature and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

SCS-Net: An efficient and practical approach towards Face Mask Detection.

Much work has been done in the computer vision domain for the problem of facial mask detection to curb the spread of the Coronavirus disease (COVID-19). Preventive measures developed using deep learning-based models have got enormous attention. With the state-of-the-art results touching perfect accuracies on various models and datasets, two very practical problems are still not addressed - the deployability of the model in the real world and the crucial cases of incorrectly worn masks. To this end, our method proposes a lightweight deep learning model with just 0.12M parameters having up to 496 times reduction as compared to some of the existing models. Our novel architecture of the deep learning model is designed for practical implications in the real world. We also augment an existing dataset with a large set of incorrectly masked face images leading to a more balanced three-class classification problem. A large collection of 25296 synthetically designed incorrect face mask images are provided. This is the first of its kind of data to be proposed with equal diversity and quantity. The proposed model achieves a competitive accuracy of 95.41% on two class classification and 95.54% on the extended three class classification with minimum number of parameters in comparison. The performance of the proposed system is assessed with various state-of-the-art literature and experimental results indicate that our solution is more realistic and rational than many existing works which use overly massive models unsuitable for practical deployability.

Masked Face Detection using Lightweight Deep Learning based Model for IoT based Healthcare applications

One of the best measures to enforce in epidemiological scenarios, such as the present COVID-19 epidemic, is the usage of masks. For a while, this will be a regular part of life, notably in public places. In order to deal with these unusual circumstances where people who wear mask are being watched, there is a need for an effective face identification technology. In order to precisely identify people wearing masks, we provide a deep learning algorithm based on YOLO architecture in this study. Unlike traditional CNNs, the proposed system uses a convergence layer to record numerous facial emotions while also using a number of convolutional filters to construct the faces for masked images. The presented design has numerous layers, including convolutional, max pooling, dropout, and softmax, and is both straightforward and effective. On the publicly accessible Real-World Masked Face Dataset, we assess the effectiveness of masked-faces detection (RWMFD). The investigational outcomes demonstrate an accurateness of 99.9%, demonstrating the effectiveness of our proposed methodology in classifying individuals wanting to wear facemasks. © 2022 IEEE.

Signal Preprocessing Technique With Noise-Tolerant for RF-Based UAV Signal Classification

Since the beginning of the COVID-19 pandemic, the demand for unmanned aerial vehicles (UAVs) has surged owing to an increasing requirement of remote, noncontact, and technologically advanced interactions. However, with the increased demand for drones across a wide range of fields, their malicious use has also increased. Therefore, an anti-UAV system is required to detect unauthorized drone use. In this study, we propose a radio frequency (RF) based solution that uses 15 drone controller signals. The proposed method can solve the problems associated with the RF based detection method, which has poor classification accuracy when the distance between the controller and antenna increases or the signal-to-noise ratio (SNR) decreases owing to the presence of a large amount of noise. For the experiment, we changed the SNR of the controller signal by adding white Gaussian noise to SNRs of -15 to 15 dB at 5 dB intervals. A power-based spectrogram image with an applied threshold value was used for convolution neural network training. The proposed model achieved 98% accuracy at an SNR of -15 dB and 99.17% accuracy in the classification of 105 classes with 15 drone controllers within 7 SNR regions. From these results, it was confirmed that the proposed method is both noise-tolerant and scalable.

Scaling Up Face Masks Classification Using a Deep Neural Network and Classical Method Inspired Hybrid Technique

Classification of persons wearing and not wearing face masks in images has emerged as a new computer vision problem during the COVID-19 pandemic. In order to address this problem and scale up the research in this domain, in this paper a hybrid technique by employing ResNet-101 and multi-layer perceptron (MLP) classifier has been proposed. The proposed technique is tested and validated on a self-created face masks classification dataset and a standard dataset. On self-created dataset, the proposed technique achieved a classification accuracy of 97.3%. To embrace the proposed technique, six other state-of-the-art CNN feature extractors with six other classical machine learning classifiers have been tested and compared with the proposed technique. The proposed technique achieved better classification accuracy and 1-6% higher precision, recall, and F1 score as compared to other tested deep feature extractors and machine learning classifiers. Copyright © 2022 KSII.

Improving COVID-19 CT classification of CNNs by learning parameter-efficient representation.

The COVID-19 pandemic continues to spread rapidly over the world and causes a tremendous crisis in global human health and the economy. Its early detection and diagnosis are crucial for controlling the further spread. Many deep learning-based methods have been proposed to assist clinicians in automatic COVID-19 diagnosis based on computed tomography imaging. However, challenges still remain, including low data diversity in existing datasets, and unsatisfied detection resulting from insufficient accuracy and sensitivity of deep learning models. To enhance the data diversity, we design augmentation techniques of incremental levels and apply them to the largest open-access benchmark dataset, COVIDx CT-2A. Meanwhile, similarity regularization (SR) derived from contrastive learning is proposed in this study to enable CNNs to learn more parameter-efficient representations, thus improve the accuracy and sensitivity of CNNs. The results on seven commonly used CNNs demonstrate that CNN performance can be improved stably through applying the designed augmentation and SR techniques. In particular, DenseNet121 with SR achieves an average test accuracy of 99.44% in three trials for three-category classification, including normal, non-COVID-19 pneumonia, and COVID-19 pneumonia. The achieved precision, sensitivity, and specificity for the COVID-19 pneumonia category are 98.40%, 99.59%, and 99.50%, respectively. These statistics suggest that our method has surpassed the existing state-of-the-art methods on the COVIDx CT-2A dataset. Source code is available at https://github.com/YujiaKCL/COVID-CT-Similarity-Regularization.

Lightweight CNN-Based Deep Neural Networks Application in Safety Measurement

Inspired by the face covering period in the past two years, COVID-19 pandemic has resulted in the mandate of public safety measures such as face mask-wearing in many countries. This paper provides a preliminary feasibility planning on how Artificial Intelligence (AI), Computer Vision (CV) and the Internet of Things (IoT) can work together to implement a face-mask detection system as a public health safety solution. This paper reviews how edge computing can overcome traditional cloud computing issues. This work also examines the current state of computer vision, convolutional neural networks and their potential application in the health and safety domain. This writing serves as an interim report on how the lightweight CNNs and single-shot detectors such as YOLOv5 variants with SSD to train and deploy an object detection system. © 2022 IEEE.

Efficient Artificial Intelligence-based Models for COVID-19 Disease Detection and Diagnosis from CT-Scans

COVID-19 is contagious virus that first emerged in China in 2019's last month. It mainly infects the both the lungs and the respiratory system. The virus has severely impacted life and the economy, which exposed threats to governments worldwide to manage it. Early diagnosis of COVID-19 could help with treatment planning and disease prevention strategies. In this study, we use CT-Scanned images of the lungs to show how COVID-19 may be identified using transfer learning model and investigate which model achieved the best and fastest results. Our primary focus was to detect structural anomalies to distinguish among COVID-19 positive, negative, and normal cases with deep learning methods. Every model received training with and without transfer learning and results were compared for various versions of DenseNet and EfficientNet. Optimal results were obtained using DenseNet201 (99.75%). When transfer learning was applied, all models produced almost similar results. © 2022 IEEE.

Human Emotion Prediction Analysis on Post-COVID-19 Crisis in Digital Media Using Deep Learning

The COVID-19 pandemic has produced a significant impact on society. Apart from its deadliest attack on human health and economy, it has also been affecting the mental stability of human being at a larger scale. Though vaccination has been partially successful to prevent further virus outreach, it is leaving behind typical health-related complications even after surviving from the disease. This research work mainly focuses on human emotion prediction analysis in post-COVID-19 period. In this work, a considerable amount of data collection has been performed from various digital sources, viz. Facebook, e-newspapers, and digital news houses. Three distinct classes of emotion, i.e., analytical, depressed, and angry, have been considered. Finally, the predictive analysis is performed using four deep learning models, viz. CNN, RNN, LSTM, and Bi-LSTM, based on digital media responses. Maximum accuracy of 97% is obtained from LSTM model. It has been observed that the post-COVID-19 crisis has mostly depressed the human being. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

An Approach of Dempster-Shafer Decision Fusion to Diagnose COVID-19 in Chest X-ray Imagery by Using Controlled Asymmetric Training of the two CNNs Ensemble

This paper proposes a model of Dempster-Shafer decision fusion based on controlled training of the ensemble of two Convolutional Neural Networks (CNNs) by the asymmetry parameter k, defined as the ratio of the numbers of training data per class assigned to each CNN module. The proposed model is dedicated to COVID-19 diagnose in chest X-ray imagery. We have considered two CNN modules with identical architectures. First CNN module has been trained with 2837 COVID-19 labeled images and (2837/k) NON-COVID images. Second CNN module has been trained with (2837/k) COVID-19 labeled images and 2837 NON-COVID images. We have evaluated the influence of control parameter k on the diagnosis performances. As a result of Dempster-Shafer fusion, for k=2.1, one obtains a maximum Overall Accuracy (OA) of 95.18% The above performance is clearly better than the corresponding OA obtained by a single CNN (92.26%) for the same k, and at the same time it is better than OA obtained by any single CNN module for any considered k. Moreover, one can remark, that by controlled training, for k=20, a CNN module can lead to an incredible low Missing Alarm Rate (MAR) of only 0.63% © 2022 IEEE.

Embedded Implementation of Social Distancing Detector Based on One Stage Convolutional Neural Network Detector

The recent COVID-19 is a very dangerous disease that intimidates humanity. It spreads very fast and many rules must be respected to reduce its prevalence. One of the most important rules is the social distance which means keeping a safe distance between two persons. A safe distance must be one meter or more. Respecting such rules in public spaces is a very challenging task that needs the assistance of artificial intelligence tools. In this paper, we propose a social distance detector using convolutional neural networks. The detector was based on the Yolo model with a custom-made backbone to guarantee real-time processing and embedded implementation. The backbone was optimized to make it suitable for embedded resources. The inference model was evaluated on the Pynq platform. The model was trained and fine-tuned using the MS COCO dataset. The evaluation of the proposed model proved its efficiency with a precision of 87.98% while running in real-time. The achieved results proved the efficiency of the proposed model and the proposed optimization for embedded implementation. © 2022 Lavoisier. All rights reserved.

Deep Learning based Analysis of Covid-19 Mortality Risk

Viral infectious diseases such as Covid-19 present a major threat to public health. Despite extreme research efforts, how, when and where such new outbreaks appear is still a source of substantial uncertainty. Deep learning (DL) is playing an increasingly important role in our lives. This paper presents one of the popular deep learning technique, Long Short Term Memory (LSTM) for prediction of Corona-Virus cases. The handcrafted feature extraction of traditional methods is less scalable on large data-sets, but deep learning algorithms perform extremely well on large data-sets, because of automatic feature extraction. Deep learning has already made a huge impact in areas, such as cancer diagnosis, precision medicine, self-driving cars, predictive forecasting, and speech recognition. This paper highlights the approaches where deep learning can be helpful to tackle the Covid-19 virus and similar outbreaks. This paper also discusses the structure and functioning of Covid-19. The utilization of different deep learning concepts like Convolutional Neural Networks, Transfer Learning for this pandemic is also highlighted. © 2022 University of Bahrain. All rights reserved.