Ensemble Image Explainable AI (XAI) Algorithm for Severe Community-Acquired Pneumonia and COVID-19 Respiratory Infections

Since the onset of the COVID-19 pandemic in 2019, many clinical prognostic scoring tools have been proposed or developed to aid clinicians in the disposition and severity assessment of pneumonia. However, there is limited work that focuses on explaining techniques that are best suited for clinicians in their decision making. In this article, we present a new image explainability method named ensemble AI explainability (XAI), which is based on the SHAP and Grad-CAM++ methods. It provides a visual explanation for a deep learning prognostic model that predicts the mortality risk of community-acquired pneumonia and COVID-19 respiratory infected patients. In addition, we surveyed the existing literature and compiled prevailing quantitative and qualitative metrics to systematically review the efficacy of ensemble XAI, and to make comparisons with several state-of-the-art explainability methods (LIME, SHAP, saliency map, Grad-CAM, Grad-CAM++). Our quantitative experimental results have shown that ensemble XAI has a comparable absence impact (decision impact: 0.72, confident impact: 0.24). Our qualitative experiment, in which a panel of three radiologists were involved to evaluate the degree of concordance and trust in the algorithms, has showed that ensemble XAI has localization effectiveness (mean set accordance precision: 0.52, mean set accordance recall: 0.57, mean set F1: 0.50, mean set IOU: 0.36) and is the most trusted method by the panel of radiologists (mean vote: 70.2%). Finally, the deep learning interpretation dashboard used for the radiologist panel voting will be made available to the community. Our code is available at https://github.com/IHIS-HealthInsights/Interpretation-Methods-Voting-dashboard. © 2020 IEEE.

COVID 19 Prediction Using Chest X-Ray And CNN Hyperparameter Optimizer

According to mid-June 2020, the abrupt escalation of coronavirus reported widespread fear and crossed 16 million confirmed cases. To fight against this growth, clinical imaging is recommended, and for illustration, X-Ray images can be applied for opinion. This paper categorizes chest X-ray images into three classes- COVID-19 positive, normal, and pneumonia affected. We have used a CNN model for analysis, and hyperparameters are used to train and optimize the CNN layers. Swarm-based artificial intelligent algorithm - Grey Wolf Optimizer algorithm has been used for further analysis. We have tested our proposed methodology, and comparative analysis has been done with two openly accessible dataset containing COVID- 19 affected, pneumonia affected, and normal images. The optimized CNN model features delicacy, insight, values of F1 scores of 97.77, 97.74, 96.24 to 92.86, uniqueness, and perfection, which are better than models at the leading edge of technology. © 2022 IEEE.

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.

Biosignal Databases for Training of Artificial Intelligent Systems

Coronavirus disease (COVID-19) is an infectious disease caused by the SARS-CoV-2 virus. Most people infected with the virus will have mild to moderate respiratory diseases, however, the elderly population is the most vulnerable, becoming seriously ill, requiring continuous medical follow-up. In this sense, technologies were developed that allow continuous and individual monitoring of patients, in a home environment, namely through wearable devices, thus avoiding continuous hospitalization. Thus, these devices allow great improvements in data analysis methods since they can continuously acquire the physiological signals of an individual and process them in real-Time through artificial intelligence (AI) methods. However, training of AI methods is not straightforward, requiring a large amount of data. In this study, we review the most common biosignal databases available in the literature. A total of thirteen databases were selected. Most of the databases (9 databases) were related to ECG signal, as well as 4 databases containing signals from SPO2, Heart Rate, Blood Pressure, etc. Characteristics were described, namely: The population of the databases, data resolution, sampling rates, sample time, number of signal samples, annotated classes, data acquisition conditions, among other aspects. Overall, this study summarizes and described the public biosignals databases available in the literature, which may be important in the implementation of intelligent classification methods. © 2022 ACM.

An artificially intelligent, natural language processing chatbot designed to promote COVID-19 vaccination: A proof-of-concept pilot study.

Objective: Our goal is to establish the feasibility of using an artificially intelligent chatbot in diverse healthcare settings to promote COVID-19 vaccination. Methods: We designed an artificially intelligent chatbot deployed via short message services and web-based platforms. Guided by communication theories, we developed persuasive messages to respond to users' COVID-19-related questions and encourage vaccination. We implemented the system in healthcare settings in the U.S. between April 2021 and March 2022 and logged the number of users, topics discussed, and information on system accuracy in matching responses to user intents. We regularly reviewed queries and reclassified responses to better match responses to query intents as COVID-19 events evolved. Results: A total of 2479 users engaged with the system, exchanging 3994 COVID-19 relevant messages. The most popular queries to the system were about boosters and where to get a vaccine. The system's accuracy rate in matching responses to user queries ranged from 54% to 91.1%. Accuracy lagged when new information related to COVID emerged, such as that related to the Delta variant. Accuracy increased when we added new content to the system. Conclusions: It is feasible and potentially valuable to create chatbot systems using AI to facilitate access to current, accurate, complete, and persuasive information on infectious diseases. Such a system can be adapted to use with patients and populations needing detailed information and motivation to act in support of their health.

Machine Learning-Based Prediction & Analysis of COVID-19 using Clinical Data

The novel corona virus (COVID-19), was initially seen in some cities of China in Dec 2019 and then spread exponentially in the entire world and converted into the worldwide pandemic. It rapidly influences and affect day to day life of everybody and slow down economy maximum countries. An immediate requirement raised to detect the positive cases on starting stage and some method to stop further spread. Radiology images have played very important role for detecting COVID-19 and it was found that these images contain very important data which is very much effective in proper diagnosis and treatment. This all creates a requirement of machine learning based artificial intelligent system to detect and further treatment of COVID-19 using X-Ray and CT images and other similar data available. Machine learning based artificial intelligent system can assist and big help for medical staff during diagnoses of COVID-19. This will also be very helpful and fill the gap of shortage of medical staff in interior towns worldwide. As we have seen that COVID-19 virus spread so fast and impact millions of patients in very short time. This creates the requirement of some computerized system that will help in diagnoses and speedy recovery of patients. One another main test which people were using was RT-PCR for detection of COVID-19 but because of many false negative results and time taken in process we need one customized Machine learning based artificial intelligent system that makes use CT images. The proposed system COVID-Rational (COVID-R) is really helpful for early detection of COVID-19 by using classification technique with supervised learning algorithms like random forest and support vector machine (SVM). We have achieved good performance assessment with accuracy of 90.2% for early detection of COVID-19 with our proposed system COVID-R. © 2022 IEEE.

A review about COVID-19 in the MENA region: environmental concerns and machine learning applications.

Coronavirus disease 2019 (COVID-19) has delayed global economic growth, which has affected the economic life globally. On the one hand, numerous elements in the environment impact the transmission of this new coronavirus. Every country in the Middle East and North Africa (MENA) area has a different population density, air quality and contaminants, and water- and land-related conditions, all of which influence coronavirus transmission. The World Health Organization (WHO) has advocated fast evaluations to guide policymakers with timely evidence to respond to the situation. This review makes four unique contributions. One, many data about the transmission of the new coronavirus in various sorts of settings to provide clear answers to the current dispute over the virus's transmission were reviewed. Two, highlight the most significant application of machine learning to forecast and diagnose severe acute respiratory syndrome coronavirus (SARS-CoV-2). Three, our insights provide timely and accurate information along with compelling suggestions and methodical directions for investigators. Four, the present study provides decision-makers and community leaders with information on the effectiveness of environmental controls for COVID-19 dissemination.

IoT and AI-enabled Physical Distance Monitoring Application to Prevent COVID19 Transmission

During COVID19 pandemic, people are encouraged to practice physical distancing at least 1 meter when interacting with other people to prevent the spread of the COVID19. This study aims to develop a system that can monitor the physical distancing and track physical contact in a room using internet of things (IoT) and artificial intelligent technology. The system consists of a small single-board computer (Raspberry Pi), webcam, and web application displaying physical contact information. The system uses YOLO algorithms to detect the human object and euclidean distance formula to determine the distance between human objects. We evaluated the performance of YOLOv3 and YOLOv3-tiny running on Raspberry Pi. The evaluation result shows that YOLOv3 consumes more CPU resources than YOLOv3-tiny but has better accuracy in detecting human objects. YOLOv3-tiny can process images and detect objects faster than YOLOv3. © 2022 IEEE.

Conclusions

This chapter presents the usage of data science, which further helps in exploring the global pandemic COVID-19. This disease suppresses an overwhelming burden, not only to healthcare systems but to the world's economy too. In this era of techniques and technologies, it is believed that data science can better utilize scarce healthcare resources. In this chapter, we provide an introduction of data science and its applications, which helps in combating different aspects of COVID-19. Publicly available datasets related to disease are used as community resources. Different kinds of datasets are used to analyze various aspects of pandemic at different scales. These different kinds of datasets can be audio, video, textual, speech, and sensor data. More than hundreds of research articles are also studied to prepare a bibliometric study. Apart from grabbing all the advantages from datasets, this paper highlights a few challenges, such as surety of correct data, need of multidisciplinary collaboration, new data modality, security issues, and availability of data. © 2021 Elsevier Inc. All rights reserved.

Artificial Intelligence in Healthcare and Medical Imaging: Role in Fighting the Spread of COVID-19

The research aims to define the need and importance of artificial intelligence in the healthcare sector in general and the medical imaging and radiology procedures in specific. This research is based on numbers and facts taking from other investigations about artificial intelligence in general and the present and studies of the healthcare sector's current Artificial Intelligence (AI) role. To simplify, the chapter will focus on the AI in healthcare facilities and how it can help solve problems and make the best decisions for the organization and public health by reducing human errors and increasing the perfection of discovering some diseases. Moreover, to achieve the medical providers' goals that aim to achieve the quality of care, we are going to talk about applying the AI in the radiology department. It's a role to help the radiologists in better diagnosis and how it could increase the efficiency of the operations by using the database of information gathered by the modern techniques. On the other hand, this article and according to the current global situation will talk about fighting the spread of Coronavirus with the new Medical Imaging technology in general and in Bahrain's society in specific.

Deep learning: A taxonomy of modern weapons to combat Covid‐19 similar pandemics in smart cities

The Covid-19 pandemic has affected many lives over the past year. In addition to the enormous health cost, the necessary lockdowns and government-mandated suspension to prevent the spread of the virus had a huge economic impact. The new challenges in 2021 were combating new virus mutations and providing effective vaccines globally. Artificial intelligent (AI) and machine learning have made significant improvements in many different applications during the last decades. One of the advanced and robust technologies in machine learning is deep learning (DL), which can be employed to help prevent initial infections and detect and monitor their progress and side effects. Fast and accurate Covid-19 infection detection and treatment of suspected patients is essential to make better decisions, ensure treatment, and even save patients' lives. Modern technologies are required to achieve these objectives and create a sustainable society. This article presents a taxonomy in DL algorithms to cover both the technical novelties and empirical results techniques for Covid-19 in smart cities. In this regard, (i) we demonstrate possible DL algorithms capable of combating Covid-19;(ii) we propose an up-to-date perspective of DL algorithms in social prevention and medical treatment;and (iii) we identify the challenges in combating Covid-19 outbreaks.

Analyzing and Forecasting Electricity Consumption During and After the Lockdown Using Nonparametric Statistical Tests and Machine Learning

The impacts of lockdown caused by the COVID-19 pandemic have dramatically changed our dealing with different lifestyles. Many countries-imposed restrictions and regulations that significantly affect the pattern of the environment and energy characterizations. In 2020, the reliability of the energy flow became crucial under lockdown situations and the changes in the electrical demand patterns. This study aims to analyze the hourly electricity usage during and after the lockdown of the COVID-19 pandemic in 2020 in Taif, Saudi Arabia and find a correlation between the lockdown and changes in consumption patterns by using nonparametric statistical tests. Moreover, a forecasting model, using the nonlinear autoregressive (NAR) artificial neural networks (ANN), is developed to predict hourly electricity consumption for a short term. The developed forecasting model considers a wide range of observations for more than 2100 hours to improve forecasting accuracy and reduce errors. This study can help decision-makers in the energy management sectors improve systems resilience and respond to shocks. © 2022 IISE Annual Conference and Expo 2022. All rights reserved.

Learning Disease Causality Knowledge From the Web of Health Data

Health information becomes importantly valuable for protecting public health in the current coronavirus situation. Knowledge-based information systems can play a crucial role in helping individuals to practice risk assessment and remote diagnosis. The authors introduce a novel approach that will develop causality-focused knowledge learning in a robust and transparent manner. Then, the machine gains the causality and probability knowledge for inference (thinking) and accurate prediction later. In addition, the hidden knowledge can be discovered beyond the existing understanding of the diseases. The whole approach is built on a causal probability description logic framework that combines natural language processing (NLP), causality analysis, and extended knowledge graph (KG) technologies. The experimental work has processed 801 diseases in total (from the UK NHS website linking with DBpedia datasets). As a result, the machine learnt comprehensive health causal knowledge and relations among the diseases, symptoms, and other facts efficiently.

DIAGNOSIS OF COVID-19 BASED ON ARTIFICIAL INTELLIGENCE MODELS AND PHYSIOLOGICAL SENSORS: REVIEW: APPLICATIONS, BASIS, AND COMMUNICATIONS

Covid-19 invaded the world very quickly and caused the loss of many lives;maximum emergency was activated all over the world due to its rapid spread. Consequently, it became a huge burden on emergency and intensive care units due to the large number of infected individuals and the inability of the medical staff to deal with patients according to the degree of severity. Covid-19 can be diagnosed based on the artificial intelligence (AI) model. Based on AI, the CT images of the patient’s chest can be analyzed to identify the patient case whether it is normal or he/she has Covid-19. The possibility of employing physiological sensors such as heart rate, temperature, respiratory rate, and SpO2 sensors in diagnosing Covid-19 was investigated. In this paper, several articles which used intelligent techniques and vital signs for diagnosing Covid-19 have been reviewed, classified, and compared. The combination of AI and physiological sensors reading, called AI-PSR, can help the clinician in making the decisions and predicting the occurrence of respiratory failure in Covid-19 patients. The physiological parameters of the Covid-19 patients can be transmitted wirelessly based on a specific wireless technology such as Wi-Fi and Bluetooth to the clinician to avoid direct contact between the patient and the clinician or nursing staff. The outcome of the AI-PSR model leads to the probability of recording and linking data with what will happen later, to avoid respiratory failure, and to help the patient with one of the mechanical ventilation devices.

Artificial Intelligence and Machine Learning Techniques for COVID-19 Prediction

An Intelligent data processing is essential to create a large amount of data in Internet of things. We progress the consistent smooth and computerized uses of artificial intelligence, machine learning, deep Learning. To analyze the data using deep learning that is subcategory of machine learning techniques. This investigation designed and implemented the intelligent system that is used to detect the rise of Covid-19 cases using various artificial intelligent algorithms through machine learning. Here best algorithm is chosen for prediction of Covid 19 Omicron cases based on their accuracy of performance metrics. © 2022 IEEE.

Semi-nested RT-PCR enables sensitive and high-throughput detection of SARS-CoV-2 based on melting analysis.

BACKGROUND: Asymptomatic transmission was found to be the Achilles' heel of the symptom-based screening strategy, necessitating the implementation of mass testing to efficiently contain the transmission of COVID-19 pandemic. However, the global shortage of molecular reagents and the low throughput of available realtime PCR facilities were major limiting factors. METHODS: A novel semi-nested and heptaplex (7-plex) RT-PCR assay with melting analysis for detection of SARS-CoV-2 RNA has been established for either individual testing or 96-sample pooled testing. The complex melting spectrum collected from the heptaplex RT-PCR amplicons was interpreted with the support of an artificial intelligence algorithm for the detection of SARS-CoV-2 RNA. The analytical and clinical performance of the semi-nested RT-PCR assay was evaluated using RNAs synthesized in-vitro and those isolated from nasopharyngeal samples. RESULTS: The LOD of the assay for individual testing was estimated to be 7.2 copies/reaction. Clinical performance evaluation indicated a sensitivity of 100% (95% CI: 97.83-100) and a specificity of 99.87% (95% CI: 99.55-99.98). More importantly, the assay supports a breakthrough sample pooling method, which makes possible parallel screening of up to 96 samples in one real-time PCR well without loss of sensitivity. As a result, up to 8,820 individual pre-amplified samples could be screened for SARS-CoV-2 within each 96-well plate of realtime PCR using the pooled testing procedure. CONCLUSION: The novel semi-nested RT-PCR assay provides a solution for highly multiplex (7-plex) detection of SARS-CoV-2 and enables 96-sample pooled detection for increase of testing capacity. .

Fuzzy Unsupervised Approaches to Analyze Covid-19 Spread for School Reopening Decision Making

Virus SARS-Cov-2 causing Covid-19 spreads quickly and brings high risks to transmissions. The government to rule strictly to arrange strategies to minimize interactions through School-From-Home (SFH) policy. Unfortunately, the school closure is the potential to hamper deliveries of education services and may entail destructive impacts to quality education performance. There must be a consideration to school reopen safely during the pandemic. The objective of the research is to produce a model of Covid-19 spreads to analyze the readiness of school to reopen. This study adopts a SEIR model to predict the spread of Covid-19 using dataset from 23 March through 31 December 2020. The best model is selected from the one having the least error and adopted to predict the spread in the next 100 days starting from 01 January 2021 through 10 April 2021. Clustering was then implemented to acquire the character's proximity in each area using K-Means algorithm. While unsupervised fuzzy was picked out to seize the phenomenon of the dynamic as Covid-19 spread as a basis to decision making on school reopen safely during the pandemic. These whole concepts will serve the decision making effectively and intelligently by generating a better estimation. This study resulted in a Covid-19 spread model with an average error of 0.2% based on the RMSLE calculation.

A Deep Neural Network for Classification of Thoracic Diseases on Chest Radiography and Covid-19 Detection

This paper presents COVID-Net, an Artificial intelligent system that can detect COVID-19 from chest X-rays based on machine learning. COVID-Net is a 3-stage machine learning (ML) system. COVID-Net is a system built on a convolutional neural network trained on over 10,000 frontal view X-ray images. The merit of this system is that it detects COVID-19 from other kinds of diseases and can be used to diagnose a new type of viral or bacterial pneumonia. © 2021 IEEE.

Predicting hospital readmission risk in patients with COVID-19: A machine learning approach.

Introduction: The Coronavirus 2019 (COVID-19) epidemic stunned the health systems with severe scarcities in hospital resources. In this critical situation, decreasing COVID-19 readmissions could potentially sustain hospital capacity. This study aimed to select the most affecting features of COVID-19 readmission and compare the capability of Machine Learning (ML) algorithms to predict COVID-19 readmission based on the selected features. Material and methods: The data of 5791 hospitalized patients with COVID-19 were retrospectively recruited from a hospital registry system. The LASSO feature selection algorithm was used to select the most important features related to COVID-19 readmission. HistGradientBoosting classifier (HGB), Bagging classifier, Multi-Layered Perceptron (MLP), Support Vector Machine ((SVM) kernel = linear), SVM (kernel = RBF), and Extreme Gradient Boosting (XGBoost) classifiers were used for prediction. We evaluated the performance of ML algorithms with a 10-fold cross-validation method using six performance evaluation metrics. Results: Out of the 42 features, 14 were identified as the most relevant predictors. The XGBoost classifier outperformed the other six ML models with an average accuracy of 91.7%, specificity of 91.3%, the sensitivity of 91.6%, F-measure of 91.8%, and AUC of 0.91%. Conclusion: The experimental results prove that ML models can satisfactorily predict COVID-19 readmission. Besides considering the risk factors prioritized in this work, categorizing cases with a high risk of reinfection can make the patient triaging procedure and hospital resource utilization more effective.

Mathematical and Engineering Solutions Facing COVID-19 Pandemic: A Review

Corona Virus disease (COVID-19) is a new disease started at December 2019 in Wuhan-China and World Health Organization declared officially that COVID-19 is a pandemic at March 2020, this pandemic is counted as the fifth one from 1918. Such situations push the researchers at every field to give their efforts to help people to get through this pandemic. Doctors, Nurses and other medical specialist start to face this situation using the knowledge they have. Beside them researchers in other fields of knowledge try to give some solutions to help the medical staff in their war against the disease. This paper is a review of the mathematical and engineering solutions that helps to know the behavior of the disease and how it spread and to give some methods to reduce the time of diagnosis the patients. © 2021 IEEE.