Data Science for COVID-19, Vol. 1: Computational Perspectives

Front Cover
Data Science for COVID-19
Data Science for COVID-19: Volume One: Computational
Copyright
Contents
Contributors
Foreword
Preface
1 - Predictive models to the COVID-19
	1. Introduction
	2. COVID-19 epidemic forecast
		2.1 Epidemic growth models
		2.2 Time series
			2.2.1 Logistic models, additive/multiplicative models, and math equations
			2.2.2 Nonlinear filter prediction models
		2.3 Machine learning prediction models
		2.4 Discussion
	3. Material and methods
		3.1 Epidemiologic predictors
			3.1.1 SIR model
			3.1.2 SEIR model
		3.2 Nonlinear additive and multiplicative methods
			3.2.1 Prophet
		3.3 Holt Winters
		3.4 Kalman filter
		3.5 State space derivation
	4. Methodology
		4.1 Performance metrics
			4.1.1 Root mean square error
			4.1.2 Mean absolute error
			4.1.3 Coefficient of determination (R2)
	5. Results
		5.1 Short-term analysis
		5.2 Long-term results for China data set
		5.3 Long-term results for hybrid data set
	6. Final considerations
	References
2 - An artificial intelligence–based decision support and resource management system for COVID-19 pandemic
	1. Introduction
	2. Fundamentals
		2.1 Mathematical epidemic models
		2.2 Decision-making in an epidemic
		2.3 Visual analysis
		2.4 Blockchain
		2.5 Data protection and privacy
	3. Related works
	4. System model
	5. Data resources
	6. Methods
		6.1 Mobile application–based tracking
		6.2 Decision support system
			6.2.1 Epidemic model
			6.2.2 Machine learning
		6.3 Blockchain
	7. Conclusion
	References
3 - Normalizing images is good to improve computer-assisted COVID-19 diagnosis
	1. Introduction
	2. Coronavirus disease 2019
	3. Proposed approach
	4. Methodology
		4.1 Datasets
		4.2 Experimental setup
	5. Experimental results
		5.1 Results
		5.2 Discussion
	6. Conclusions and future works
	Acknowledgments
	References
4 - Detection and screening of COVID-19 through chest computed tomography radiographs using deep neural networks.
	1. Introduction
	2. Symptoms and characteristics of COVID-19
	3. Screening for COVID-19
		3.1 Polymerase chain reaction
		3.2 Radiography
	4. Deep model for COVID-19 detection
		4.1 Data acquisition
	5. Preprocessing
	6. Experiment
		6.1 Model architecture
		6.2 Training and results
	7. Discussion and conclusion
	8. Current research and future work
	References
5 - Differential evolution to estimate the parameters of a SEIAR model with dynamic social distancing: the case of COVID-19 ...
	1. Introduction
	2. Related works
	3. The SEIAR model
		3.1 The SEIAR model with social distancing
		3.2 The SEIAR–SD model with dynamic social distancing
	4. Simulations
		4.1 Methods
		4.2 Simulation results
			4.2.1 Modeling Italy
			4.2.2 Modeling Lombardy
			4.2.3 Modeling Veneto
			4.2.4 Modeling Emilia–Romagna
			4.2.5 Modeling Campania
		4.3 Discussion
	5. Conclusions
	References
6 - Limitations and challenges on the diagnosis of COVID-19 using radiology images and deep learning
	1. Introduction
	2. COVID-19 radiology imaging dataset
	3. Recent works using radiology images for COVID-19
	4. Deep learning basics
		4.1 Convolutional neural networks
		4.2 EfficientNet as the deep learning model
		4.3 Using EfficientNet as classification method for radiology images
	5. Limitations and challenges
		5.1 Small data
		5.2 Explainability
		5.3 The perspective of differential privacy
	6. Summary and future perspective
	Acknowledgments
	References
	Further reading
7 - Deep convolutional neural network–based image classification for COVID-19 diagnosis
	1. Introduction
		1.1 Problem statement
		1.2 Objective
	2. Overview of data processing
		2.1 Statistical methods for data processing
	3. Overview on COVID-19 datasets
		3.1 Numeric data for COVID-19 diagnosis
		3.2 Medical data for COVID-19 diagnosis
	4. Background study
	5. Proposed system for COVID-19 detection using image classification
		5.1 Convolutional neural network architecture
			5.1.1 Input layer
			5.1.2 Convolution layer
			5.1.3 Pooling layer
			5.1.4 Activation function
			5.1.5 Fully connected layer
	6. Materials and methods
		6.1 Datasets
		6.2 Learning frameworks
		6.3 Preprocessing of datasets
			6.3.1 Median filter
			6.3.2 Wiener filter
			6.3.3 Histogram equalization
			6.3.4 Adaptive histogram equalization and contrast limited adaptive histogram equalization
			6.3.5 Gamma correction
		6.4 Data preparation and augmentation
	7. Model training
		7.1 Feature extraction
			7.1.1 DenseNet169
			7.1.2 Channel boosted convolutional neural network
			7.1.3 ResNeXt
			7.1.4 AlexNet
			7.1.5 VGG16
		7.2 Classification
	8. Results and discussions
	9. Conclusion
	References
8 - Statistical machine learning forecasting simulation for discipline prediction and cost estimation of COVID-19 pandemic
	1. Introduction
	2. Literature survey of situation report by World Health Organization
	3. Supervised model for discipline analysis within a country against COVID-19
		3.1 Statistical machine learning forecasting strategy
	4. Limitations and future scope
	References
	Appendix I. Prediction data of top 15 countries
	Appendix II. Prediction data of India and subcontinental countries
9 - Application of machine learning for the diagnosis of COVID-19
	1. Introduction
	2. Visualization of the spread of coronavirus disease 2019
	3. Methodology
	4. Feature importance and feature scoring
	5. Classification using machine learning
		5.1 XGBoost
		5.2 Random forest
		5.3 Decision tree and logistic regression
	6. Performance parameters
	7. Conclusion
	References
10 - PwCOV in cluster-based web server: an assessment of service-oriented computing for COVID-19 disease processing system
	1. Introduction
	2. Materials and method
		2.1 Overview of the system metrics
		2.2 Related work
	3. Focus of the study
	4. Testing of PwCOV
		4.1 Distribution of data points for 50 virtual user
		4.2 Correlation of system metrics
	5. Reliability of PwCOV
	6. Overall assessment of PwCOV
	7. Conclusion
	Acknowledgment
	References
11 - COVID-19–affected medical image analysis using DenserNet
	1. Introduction
	2. Related works
	3. Problem formulation
	4. Proposed methodology
	5. Experiments and discussions
		5.1 Database employed
		5.2 Experimental results
		5.3 Comparison
		5.4 Research impact
	6. Conclusion
	References
12 - uTakeCare: unlock full decentralization of personal data for a respectful decontainment in the context of COVID-19: to ...
	1. Introduction
	2. COVID-19 public safety applications
	3. Ethical and legal discussion on COVID-19 digital applications
	4. uTakeCare: a new concept of deconfinement applications
		4.1 General idea
		4.2 Existing methods for estimating COVID-19 vulnerability
		4.3 uTakeCare COVID-19 vulnerability estimator
			4.3.1 Training stage
			4.3.2 Computation stage
		4.4 Zero-knowledge proof as a solution to obtain full anonymization?
			4.4.1 Application scenarios
	5. Limitations, perspectives, and futures works
	6. Conclusion
		6.1 Source code policy
	Acknowledgments
	References
13 - COVID-19 detection from chest X-rays using transfer learning with deep convolutional neural networks
	1. Introduction
		1.1 Literature review
	2. Materials and method
		2.1 X-ray data
		2.2 Pretrained deep convolutional neural network models
			2.2.1 VGGNet
			2.2.2 MobileNet
			2.2.3 ResNets
			2.2.4 DenseNet
	3. Experimental results
		3.1 Dataset-A
		3.2 Dataset-B
	4. Conclusion
	References
14 - Lexicon-based sentiment analysis using Twitter data: a case of COVID-19 outbreak in India and abroad
	1. Introduction
	2. Proposed methodology
		2.1 Data collection
			2.1.1 Global Twitter data
			2.1.2 Indian Twitter data
		2.2 Data preprocessing
		2.3 Sentiment analysis approach
			2.3.1 Lexicon-based approach
			2.3.2 VADER-based approach
		2.4 Sentiment analysis
			2.4.1 Result of Lexicon-based approach
			2.4.2 Result of VADER-based approach
			2.4.3 Sentiment analysis of India during and before lockdown
	3. Discussion
	4. Conclusion
	References
15 - Real-time social distance alerting and contact tracing using image processing
	1. Introduction
	2. Flattening the curve
		2.1 Susceptible, exposed, infected, and recovered model
	3. Contact tracing
	4. Proposed system for identification of susceptible members
		4.1 Identification of unique people in the frame
			4.1.1 SSD—single shot detection
			4.1.2 MobileNet single shot detection
			4.1.3 YOLO—you only look once
			4.1.4 Implementation
		4.2 Susceptibility score as a concept for contact tracing
	5. Conclusion
	References
16 - Machine-learning models for predicting survivability in COVID-19 patients
	1. Introduction
	2. Materials and method
		2.1 A prediction of survivability of the COVID-19 patients using machine learning
			2.1.1 Data collection
			2.1.2 Data preprocessing
			2.1.3 Feature selection
			2.1.4 The machıne learnıng models
				2.1.4.1 Decision tree
				2.1.4.2 Random forest
				2.1.4.3 Logistic regression
				2.1.4.4 Gradient boosting
	3. Comparative analysis and results
		3.1 Evaluation metrics
	4. Discussion
	5. Conclusion
	References
17 - Robust and secured telehealth system for COVID-19 patients
	1. Introduction
		1.1 Big data and pandemic
			1.1.1 Artificial intelligence
		1.2 Critical data transmission in telehealth
			1.2.1 Error correction mechanisms
	2. Error mitigation codes for telehealth system
		2.1 New reduced complexity low-density parity-check Min-Sum decoder
			2.1.1 Simulation results
	3. Conclusion
	References
18 - A novel approach to predict COVID-19 using support vector machine
	1. Introduction
	2. Related studies
	3. Proposed COVID-19 detection methodology
		3.1 Case 1: not infected
		3.2 Case 2: mildly infected
		3.3 Case 3: severely infected
	4. Experimental results and discussions
	5. Performance analysis of other supervised learning models using visual programming
	6. Concluding remarks
	References
19 - An ensemble predictive analytics of COVID-19 infodemic tweets using bag of words
	1. Introduction
	2. Related models
	3. Research methodology
		3.1 Tweet-data acquisition phase
		3.2 Tweet Tokenization and information gain evaluation
		3.3 Singly and ensemble classification stage
	4. Results and discussion
	5. Conclusion and recommendation
	Acknowledgments
	References
20 - Forecast and prediction of COVID-19 using machine learning
	1. Introduction
	2. Introduction to COVID-19
		2.1 Incubation period of COVID-19
		2.2 How it is transmitted
		2.3 Symptoms of COVID-19
		2.4 Countries most affected by COVID-19
	3. Introduction to machine learning
		3.1 Some machine learning methods
	4. Use of machine learning in COVID-19
	5. Different techniques for prediction and forecasting
	6. Proposed method for prediction
	7. Forecasting
	8. Conclusion and future work
	References
21 - Time series analysis of the COVID-19 pandemic in Australia using genetic programming
	1. Introduction
	2. Technical preliminaries and model calibration
		2.1 Gene expression programming
		2.2 Proposed gene expression programming model
	3. Proposed gene expression programming–based formulation for best OBJ
	4. Model validity and comparative study
	5. Variable importance
	6. Conclusion
	References
22 - Image analysis and data processing for COVID-19
	1. Introduction
	2. Explanations regarding detection and analysis for COVID-19
		2.1 Data augmentation technique to analyze COVID-19 patients
		2.2 Image analysis for detection of COVID-19
	3. Data processing to analyze the number of COVID-19 patients
		3.1 Prediction of treatment consequences for COVID-19
		3.2 Recognition of existing drugs and drug development
	4. Explanation of patient chest computed tomography scan imaging analysis using deep learning
	5. Conclusion
	References
23 - A demystifying convolutional neural networks using Grad-CAM for prediction of coronavirus disease (COVID-19) on X-ray  ...
	1. Introduction
	2. Literature survey
		2.1 Objective of the work
		2.2 Vital opinions
	3. Materials and method
		3.1 Dataset introduction
		3.2 Dataset description
			3.2.1 Data working model explanation
			3.2.2 Data augmentation
	4. Implementation workflow
		4.1 Training the data with state-of-the-art models
			4.1.1 Results of the model
		4.2 Deep transfer learning with layer fine turning and VGG 16
			4.2.1 Experimental setup for training the model carried out
		4.3 Training with a convolutional neural network architecture
			4.3.1 Performance metrics
	5. Gradient-based activation model
	6. Results discussion
	7. Conclusion
	8. Future work
	9. Summary of work carried out so far
	10. Application program interface for COVID-19 testing
		Note
		Author contributions statement
		Additional information
	Acknowledgments
	References
24 - Transfer learning-based convolutional neural network for COVID-19 detection with X-ray images
	1. Introduction
	2. Convolutional neural network
		2.1 Transfer learning
			2.1.1 Visual Geometry Group Network (VGG)
			2.1.2 ResNet
			2.1.3 DenseNet
			2.1.4 InceptionV3
	3. Materials and method
		3.1 Dataset
		3.2 Evaluation
		3.3 Proposed model
	4. Conclusion
	References
25 - Computational modeling of the pharmacological actions of some antiviral agents against SARS-CoV-2
	1. Introduction
	2. Material and methods
		2.1 Proteins and ligands collection
		2.2 Absorption, distribution, metabolism, excretion, and toxicity screening
		2.3 Virtual screening
		2.4 Statistical analysis
	3. Results
		3.1 Binding affinity of the antiviral drugs and their derivatives with SARS-CoV-2 target proteins
		3.2 Absorption, distribution, metabolism, excretion, and toxicity predictions
		3.3 Toxicity profile of potential SARS-CoV-2 inhibitors
		3.4 Molecular docking interactions
	4. Discussion
	5. Conclusion
	References
	Further reading
26 - Received signal strength indication—based COVID-19 mobile application to comply with social distancing using bluetooth ...
	1. Introduction
	2. Literature review
		2.1 Other solutions for social distancing using other technologies
		2.2 Indoor localization systems and its relation with Received Signal Strength Indication
		2.3 Factors that affect the measurements of Received Signal Strength Indication
		2.4 Received Signal Strength Indication filtering techniques
	3. Experiment overview
		3.1 Methodology for measurements
		3.2 Collection of data
	4. Analysis of results
		4.1 Descriptive analysis
		4.2 Machine learning model
		4.3 Embedding a machine learning model in a smartphone using TensorFlow Lite
		4.4 Operation of the application
	5. Discussion
	6. Conclusions and future work
	References
27 - COVID-19 pandemic in India: forecasting using machine learning techniques
	1. Introduction
	2. Material and methods
		2.1 Dataset
		2.2 Preprocessing
			2.2.1 Normalization
			2.2.2 Sliding window
		2.3 Feature extraction and feature selection
	3. Machine learning techniques
	4. Results and discussion
		4.1 Experimental design
		4.2 Regression-based model development
		4.3 Performance analysis
			4.3.1 Model validation
			4.3.2 Model analysis by plotting
			4.3.3 N-days-ahead forecasting of novel coronavirus disease 2019 pandemic
	5. Conclusion
	References
28 - Mathematical recipe for curbing coronavirus (COVID-19) transmition dynamics
	1. Introduction
	2. Materials and methods
	3. Proposed model
	4. Existence and uniqueness of solution of the model
	5. Stability analysis (positivity solution)
	6. Model equilibrium point
	7. Results
		7.1 Effective reproduction number (Ro)
		7.2 Existence of endemic equilibrium
		7.3 Local stability of the model
		7.4 The characteristics equation is given as follows
		7.5 Numerical simulation results and discussion
	8. Discussion
	9. Conclusion
	Acknowledgments
	References
29 - Sliding window time series forecasting with multilayer perceptron and multiregression of COVID-19 outbreak in Malaysia
	1. Introduction
	2. Related work
		2.1 Time series forecasting for coronavirus disease 2019
		2.2 Data mining and artificial intelligence solutions combatting coronavirus disease 2019
	3. Sliding window technique for temporal data analytics
	4. Trend analysis and forecast
		4.1 The overall data
		4.2 Trend of cumulative coronavirus disease 2019 cases in Malaysia
		4.3 Sliding window time series forecasting for coronavirus disease 2019
		4.4 Data preparation with sliding window representation
			4.4.1 Sliding window time series forecasting with multiple regression
			4.4.2 Sliding window time series forecasting with multilayer perceptron with lag
	5. Discussion
	6. Conclusion
	Acknowledgments
	References
30 - A two-level deterministic reasoning pattern to curb the spread of COVID-19 in Africa
	1. Introduction
	2. Proposed two-level deterministic reasoning pattern for COVID-19
		2.1 Rescue-to-discharge pattern for COVID-19
		2.2 Transition Chain Petri net
		2.3 Petri net model to determine death rate in COVID-19 cases
	3. Determining distribution function for Petri net with COVID-19 cases
		3.1 Discussion
	4. Conclusion
	References
31 - Data-driven approach to COVID-19 infection forecast for Nigeria using negative binomial regression model
	1. Introduction
	2. Material and methods
		2.1 Data
		2.2 Preprocessing
		2.3 Prediction model
	3. Results and discussion
		3.1 Goodness of fit test
		3.2 Model effect and statistical analysis of predictor variables
			3.2.1 Statistical significance of predictor variables
			3.2.2 Model effect of predictor variables
		3.3 Prediction accuracy
	4. Conclusion
	Acknowledgments
	References
32 - A novel machine learning–based detection and diagnosis model for coronavirus disease (COVID-19) using discrete wavelet ...
	1. Introduction
	2. The proposed discrete wavelet transform–rough neural network model
		2.1 Discrete wavelet transform–based feature extraction
			2.1.1 Discrete wavelet transform
			2.1.2 Two-dimensional discrete wavelet transform
		2.2 Polymerase chain reaction–based feature reduction
		2.3 Rough neural network–based classification
			2.3.1 Level 1: preprocessing
			2.3.2 Level 2: training
			2.3.3 Level 3: testing phase
	3. Performance validation
		3.1 Dataset description
		3.2 Results of analysis
	4. Conclusion
	Acknowledgment
	References
33 - Artificial intelligence–based solutions for early identification and classification of COVID-19 and acute respiratory  ...
	1. Introduction
	2. The proposed enhanced kernel support vector machine model
		2.1 Preprocessing
		2.2 Feature extraction using Hough transform
		2.3 Particle swarm optimization–kernel support vector machine classifier
			2.3.1 Support vector machine classifier
				2.3.1.1 Principles of linear support vector machines
				2.3.1.2 Soft margin
				2.3.1.3 Dual form
			2.3.2 Kernel support vector machine
			2.3.3 Particle swarm optimization–kernel support vector machine
		2.4 Cross-validation
	3. Experimental validation
	4. Conclusion
	References
34 - Internet of Medical Things (IoMT) with machine learning–based COVID-19 diagnosis model using chest X-ray images
	1. Introduction
	2. The proposed model
		2.1 Gray-level co-occurrence matrix (GLCM)-based feature extraction
			2.1.1 Angular second moment
			2.1.2 Contrast
			2.1.3 Inverse difference moment
			2.1.4 Entropy
			2.1.5 Correlation
			2.1.6 Sum of square and variance
			2.1.7 Difference entropy
			2.1.8 Inertia
			2.1.9 Cluster shade
			2.1.10 Cluster prominence
			2.1.11 Energy
			2.1.12 Homogeneity
			2.1.13 Dissimilarity
			2.1.14 Difference in variance
		2.2 Classification models
			2.2.1 Support vector machine model
			2.2.2 Artificial neural network classifier
			2.2.3 Decision tree classifier
			2.2.4 AdaBoost with random forest model
	3. Performance validation
	4. Conclusion
	References
35 - The growth of COVID-19 in Spain. A view based on time-series forecasting methods
	1. Introduction
	2. Materials and method
		2.1 Time-series models
			2.1.1 ARIMA models
			2.1.2 Cross-correlations
		2.2 Data sources
	3. Analysis of the daily death toll
	4. Analysis of the relationship between deaths and intensive care unit figures
	5. Relationship between infected and recovered
	6. Conclusions and final comments
	Annex A. Data
	References
36 - On privacy enhancement using u-indistinguishability to COVID-19 contact tracing approach in Korea
	1. Introduction
	2. Related technologies
	3. Contact tracing in South Korea
	4. Problems of contact data disclosure
	5. u-indistinguishability
	6. Conclusion
	Acknowledgment
	References
37 - Scheduling shuttle ambulance vehicles for COVID-19 quarantine cases, a multi-objective multiple 0–1 knapsack model wit ...
	1. Introduction
	2. Scheduling shuttle ambulance for COVID-19 patients
	3. Multi-objective Multiple Knapsack Problem: an overview
	4. Mathematical model for scheduling the shuttle ambulance vehicles
		4.1 Mathematical model
		4.2 Constraints
	5. An illustrated case study
	6. Proposed methodology
		6.1 Gaining sharing-knowledge-based optimization algorithm
		6.2 Discrete Binary Gaining-Sharing knowledge-based optimization algorithm
			6.2.1 Discrete binary initialization
			6.2.2 Discrete binary junior gaining and sharing stage
			6.2.3 Discrete binary senior gaining and sharing stage
	7. Experimental results
	8. Conclusions and points for future researches
	References
Index
	A
	B
	C
	D
	E
	F
	G
	H
	I
	J
	K
	L
	M
	N
	O
	P
	Q
	R
	S
	T
	U
	V
	W
	X
	Y
	Z
Back Cover