Machine Learning and Deep Learning in Efficacy Improvement of Healthcare Systems

Description

The goal of medical informatics is to improve life expectancy, disease diagnosis and quality of life. Medical devices have revolutionized healthcare and have led to the modern age of machine learning, deep learning and Internet of Medical Things (IoMT) with their proliferation, mobility and agility. This book exposes different dimensions of applications for computational intelligence and explains its use in solving various biomedical and healthcare problems in the real world. This book describes the fundamental concepts of machine learning and deep learning techniques in a healthcare system. The aim of this book is to describe how deep learning methods are used to ensure high-quality data processing, medical image and signal analysis and improved healthcare applications. This book also explores different dimensions of computational intelligence applications and illustrates its use in the solution of assorted real-world biomedical and healthcare problems. Furthermore, it provides the healthcare sector with innovative advances in theory, analytical approaches, numerical simulation, statistical analysis, modelling, advanced deployment, case studies, analytical results, computational structuring and significant progress in the field of machine learning and deep learning in healthcare applications. FEATURES Explores different dimensions of computational intelligence applications and illustrates its use in the solution of assorted real-world biomedical and healthcare problems Provides guidance in developing intelligence-based diagnostic systems, efficient models and cost-effective machines Provides the latest research findings, solutions to the concerning issues and relevant theoretical frameworks in the area of machine learning and deep learning for healthcare systems Describes experiences and findings relating to protocol design, prototyping, experimental evaluation, real testbeds and empirical characterization of security and privacy interoperability issues in healthcare applications Explores and illustrates the current and future impacts of pandemics and mitigates risk in healthcare with advanced analytics This book is intended for students, researchers, professionals and policy makers working in the fields of public health and in the healthcare sector. Scientists and IT specialists will also find this book beneficial for research exposure and new ideas in the field of machine learning and deep learning. Cover Half Title Series Page Title Page Copyright Page Contents Preface Editors Contributors 1. Machine Learning in Healthcare: An Introduction 1.1 Introduction 1.2 Machine Learning and Its Basic Workings 1.3 Why Machine Learning? 1.4 Machine Learning Techniques 1.4.1 Supervised Learning 1.4.1.1 Workings of Support Vector Machine Algorithm 1.4.2 Unsupervised Learning 1.4.2.1 Clustering Types 1.5 Understanding the Healthcare Industry 1.6 Applications of ML in Healthcare 1.6.1 Machine Learning in Prognosis 1.6.2 Machine Learning in Diagnosis 1.6.3 Electronic Health Records 1.6.3.1 Electronic Health Records and Machine Learning 1.6.4 Applications of ML in Medical Image Analysis 1.6.5 Machine Learning in Natural Language Processing of Medical Documents and Literature 1.6.6 Machine Learning and Pandemic Combatting 1.6.7 Applications of ML in Pandemic Predictions 1.6.8 Applications of ML in Pandemic Controls 1.7 Conclusion References 2. A Machine Learning Approach to Identify Personality Traits from Social Media 2.1 Introduction 2.2 Related Works 2.3 Proposed Methodology 2.3.1 Classification Using Random Forest 2.3.2 KNN or K-Nearest Neighbour 2.3.3 SVM or Support Vector Machine 2.3.4 Naïve-Bayes 2.3.5 Long Short Term Memory or LSTM 2.3.6 Convolutional Neural Networks or CNN 2.4 Experimental Result and Performance Evaluation 2.4.1 Data Set Preparation 2.4.1.1 Data Set Visualisation 2.4.2 Pre-processing of Data 2.4.3 Classification 2.4.3.1 Random Forest 2.4.3.2 K-Nearest Neighbour 2.4.3.3 Support Vector Machine 2.4.3.4 Naïve-Bayes 2.4.3.5 Convolutional Neural Network 2.4.3.6 Long Short Term Memory 2.5 Results 2.6 Future Scope 2.7 Conclusion References 3. Influence of Content Strategies on Community Engagement over the Healthcare-Related Social Media Pages in India 3.1 Introduction 3.2 Literature Review 3.2.1 Social Media 3.2.2 Social Media and Healthcare 3.2.3 Engagement over Social Media 3.3 Method 3.3.1 Variables Operationalization 3.3.2 Model Specification 3.4 Results 3.4.1 Results of Poisson Regression 3.5 Conclusion 3.6 Limitations References 4. The Impact of Social Media in Fighting Emerging Diseases: A Model-Based Study 4.1 Introduction 4.2 Literature Review 4.3 The Mathematical Model 4.4 Some Preliminary Results 4.4.1 Equilibria 4.4.1.1 Disease-Free Equilibrium 4.4.1.2 Endemic Equilibrium 4.4.2 Hopf Bifurcation at Coexistence 4.5 Numerical Simulations 4.5.1 System Behaviour Changes for A 4.5.2 Dynamical Changes due to k and β 4.5.3 Impact of A on the Infected Population 4.5.4 Impact of k on the Infected Population 4.5.5 Effect of β 4.5.6 Effect of λ 4.5.7 Impact of Treatment (b) on the Infected Population 4.5.8 Two-Parameter Bifurcation Diagram 4.6 Discussion 4.7 Conclusion References 5. Prediction of Diabetes Mellitus Using Machine Learning 5.1 Introduction 5.2 Machine Learning 5.2.1 Supervised Learning 5.2.2 Unsupervised Learning 5.2.3 Reinforcement Learning 5.3 Literature Review 5.4 Workflow 5.5 Proposed Framework Model 5.6 Methods of Classification and Evaluations 5.6.1 Support Vector Machine 5.6.2 Naïve-Bayes 5.6.3 Decision Tree (DT) 5.7 Symptoms of Type 1 Diabetes and Type 2 Diabetes 5.7.1 The Risk Factors for Type 1 and Type 2 Diabetes 5.8 Results and Discussion 5.9 Conclusion and Future Work References 6. Spectrogram Image Textural Descriptors for Lung Sound Classification 6.1 Introduction 6.2 Literature Survey 6.3 Proposed Approach for Lung Sound Classification Using Time-Frequency Textural Features 6.4 Pre-processing Techniques 6.4.1 Conventional Spectrogram 6.4.2 Log-Mel Spectrogram 6.4.3 Constant-Q Transform (CQT) 6.5 Feature Extraction, Feature Selection, and Classification 6.5.1 Local Binary Pattern (LBP) 6.5.2 Completed Local Binary Pattern (CLBP) 6.5.3 Local Phase Quantization (LPQ) 6.5.4 Neighbourhood Component Analysis (NCA) 6.5.5 Decision Tree 6.6 Experimental Results and Discussion 6.6.1 Database 6.6.2 Pre-processing 6.6.3 Feature Extraction and Feature Selection 6.6.4 Classification Using Decision Tree and Performance Evaluation 6.7 Conclusion References 7. Medical Image Analysis Using Machine Learning Techniques: A Systematic Review 7.1 Introduction 7.2 Methodology 7.3 History and Characteristics of Medical Images 7.4 Machine Learning Application in Medical Imaging 7.4.1 Artificial Neural Network 7.4.1.1 Analysis of Previous Methods 7.4.1.2 Proposed Solutions 7.4.1.3 Results 7.4.2 K-Nearest Neighbour Algorithm 7.4.2.1 Analysis of Previous Methods 7.4.2.2 Proposed Solutions 7.4.2.3 Results 7.4.3 Genetic Algorithm 7.4.3.1 Analysis of Previous Methods 7.4.3.2 Proposed Solutions 7.4.3.3 Results 7.4.4 Ant Community Optimization 7.4.4.1 Analysis of Previous Methods 7.4.4.2 Proposed Solutions 7.4.4.3 Results 7.5 Discussion and Conclusions References 8. Impact of Ensemble-Based Models on Cancer Classification, Its Development, and Challenges 8.1 Introduction 8.2 Types of Ensembles and Their Application on Cancer Classification 8.3 Material and Methods 8.3.1 Data Set Description 8.3.2 Maximum Relevance Minimum Redundancy (MRMR) 8.3.3 Support Vector Machine 8.3.4 Decision Tree 8.3.5 Naïve-Bayes 8.3.6 Neural Network 8.3.7 Logistic Regression 8.3.8 KNN 8.3.9 Proposed Stacking Ensemble Model 8.4 Experimental Results and Discussion 8.4.1 Performance Measure 8.4.2 Results and Discussion 8.5 Conclusion References 9. Performance Comparison of Different Machine Learning Techniques towards Prevalence of Cardiovascular Diseases (CVDs) 9.1 Introduction 9.2 Literature Review 9.3 Data Pre-processing 9.4 Proposed Methodologies 9.4.1 Support Vector Machine (SVM) 9.4.2 Naïve-Bayes (NB) 9.4.3 Logistic Regression (LR) 9.4.4 Bayesian Regularization Neural Network (BRNN) 9.5 Experimental Results 9.6 Conclusion and Future Scopes of the Study References 10. Deep Neural Networks in Healthcare Systems 10.1 Introduction 10.2 COVID-19 Evolution and Emergence 10.2.1 COVID-19 Situation Worldwide 10.2.2 COVID-Situation in Bharat 10.3 COVID-19 Detection and Measurement History 10.4 AI in COVID-19 Disease 10.4.1 Artificial Intelligence 10.4.2 Supervised Learning 10.4.3 Unsupervised Learning 10.4.4 Reinforcement Learning 10.5 AI Applications in Fighting against COVID-19 10.5.1 Detection and Diagnosis of COVID-19 10.5.2 Identifying, Tracking, and Predicting the Outbreak 10.5.3 AI for Infodemiology and Infoveillance 10.5.4 AI for Biomedicine and Pharmacotherapy 10.6 Metrics for Deep Learning-Based Analysis 10.7 COVID-19 Detection Techniques 10.7.1 Use of Chest X-rays and CT Images for COVID-19 10.7.2 Image-Based Diagnosis of COVID-19 Using ML 10.7.3 X-Ray Images Utilizing Transfer Learning with CNN 10.7.4 Computer Vision and Radiology for COVID-19 Detection 10.8 Forecasting Models 10.8.1 Big Data 10.8.2 Social Media Data/Other Communication Media Data 10.8.3 Stochastic Theory/Mathematical Models 10.8.4 Data Science and ML Techniques 10.9 Data-Driven Analytical Models of COVID-19 10.9.1 Exponential Model 10.9.2 Logistic Model 10.9.3 SIR Model 10.9.4 MetaWards 10.9.5 SIDARTHE 10.10 Visualizing Trends of COVID-19 10.11 Challenges in Data Collection 10.11.1 Regulation 10.11.2 Data Inefficiency 10.11.3 Privacy 10.12 Conclusion and Future Scope References 11. Deep Learning and Multimodal Artificial Neural Network Architectures for Disease Diagnosis and Clinical Applications 11.1 Introduction 11.2 Technologies in Healthcare Sector 11.3 Applications of Artificial Intelligence in Healthcare 11.4 Machine Learning Techniques in the Healthcare Sector 11.4.1 Naïve-Bayes Classifier 11.4.2 Support Vector Machines 11.4.3 Decision Trees 11.4.4 Random Forest Classifier 11.4.5 Artificial Neural Networks 11.5 Deep Learning Approach in Healthcare 11.6 Deep Neural Network Architectures in the Medical Field 11.6.1 Convolutional Neural Network 11.6.2 Restricted Boltzmann Machine 11.6.3 Deep Belief Networks 11.6.4 Autoencoder 11.6.5 Recurrent Neural Networks 11.6.6 Long Short Term Memory 11.6.7 Generative Adversarial Nets 11.7 Hybrid Approach of Machine Learning and Deep Learning Techniques 11.8 Mathematical Model of Multimodal Neural Network for Disease Prediction and Labeling 11.9 Conclusion References 12. A Temporal JSON Model to Represent Big Data in IoT-Based e-Health Systems 12.1 Introduction 12.2 Related Work 12.2.1 IoT, IoMT and IoT-Based e-Health Systems 12.2.2 Machine Learning and Deep Learning in Healthcare Systems 12.2.3 Big Data Modeling 12.2.4 Temporal Database Concepts 12.3 TJeH: Our Temporal JSON Model for e-Health IoT Data 12.3.1 Running Example 12.4 C-TJeH: Our Graphical Conceptual Model for Temporal JSON e-Health IoT Data 12.4.1 Conceptual Modeling of Conventional Aspects of e-Health IoT Data under C-TJeH 12.4.2 Conceptual Modeling of Temporal Aspects of e-Health IoT Data under C-TJeH 12.4.3 Running Example Reprise 12.5 Conclusion References 13. Use of UAVs in the Prevention, Control and Management of Pandemics 13.1 Introduction 13.1.1 Chapter Organization 13.2 The Use of Drones in the Pandemic Emergency: Pandemic Drones 13.3 Materials and Methods 13.3.1 Delivering Medical Goods 13.3.2 Monitoring Crowd and Flows 13.3.3 Machine Learning Algorithms and Codes Used for the Proposed Automated System 13.3.3.1 IFTTT (If This Then That) 13.3.3.2 YOLO 13.3.3.3 SORT 13.3.3.4 Backpropagation 13.3.3.5 MAC Address 13.3.3.6 Genetic Algorithm 13.4 Case Study 13.4.1 Medical Goods Delivery 13.4.2 Monitoring and Control of Movements and Crowding 13.5 Discussion 13.6 Conclusions Notes References 14. Implicit Ontology Changes Driven by Evolution of e-Health IoT Sensor Data in the τOWL Semantic Framework 14.1 Introduction 14.2 Related Work 14.2.1 The τOWL Framework 14.2.2 Ontology-Based IoT e-Health Systems 14.2.3 Machine Learning and Deep Learning in Healthcare Systems 14.3 Implicit Ontology Structure Changes Triggered by Non-Conservative Updates to Ontology Data Instances 14.4 Extending τOWL to Support Non-Conservative Updates of Ontology Instances 14.5 Proof-of-Concept Extension of the τOWL-Manager Tool 14.6 Conclusion Notes References 15. Classification of Text Data in Healthcare Systems - A Comparative Study 15.1 Introduction 15.2 Related Work 15.3 Text Classification Algorithms and Techniques 15.3.1 Text Classification Framework 15.3.2 Data Correction 15.3.3 Data Pre-processing 15.3.4 Vectorization of Text Data 15.3.4.1 Bag-of-Words Approach 15.3.4.2 Vector Weighting Techniques 15.3.4.3 Term Frequency (TF) 15.3.4.4 Term Frequency - Inverse Document Frequency (TF-IDF) 15.3.5 Word Embeddings 15.3.5.1 Fasttext 15.3.6 Sampling Methods 15.3.6.1 N-Fold Cross-Validation (N-Fold CV) 15.3.7 Supervised Machine Learning-Based Classifiers 15.3.7.1 Naïve-Bayes (NB) 15.3.7.2 Logistic Regression (LR) 15.3.7.3 Support Vector Machine (SVM) 15.3.7.4 K-Nearest Neighbors (KNN) 15.3.7.5 Decision Tree (DT) 15.3.7.6 Random Forest (RF) 15.3.8 Deep Learning-Based Classifiers 15.3.8.1 Artificial Neural Networks (ANNs) 15.3.9 Pre-Trained Language Models 15.3.9.1 Bidirectional Encoder Representations from Transformers (BERT) 15.3.10 Evaluation Metrics 15.3.10.1 Accuracy 15.3.10.2 Precision 15.3.10.3 Recall 15.3.10.4 F-Measure 15.4 Research Methodology 15.5 Data Set 15.6 Experiments and Results 15.6.1 Computing Setup 15.6.2 Experiments Using Supervised Machine Learning Algorithms 15.6.3 Experiments Using Word Embeddings 15.6.4 Experiments Using Pre-Trained Language Models 15.7 Discussion and Conclusion References 16. Predicting Air Quality Index with Machine Learning Models 16.1 Introduction 16.2 Related Work 16.3 Model Selection 16.4 Regression Models 16.4.1 Multiple Linear Regression 16.4.2 Decision Tree Regression 16.4.3 Random Forest Regression 16.4.4 Support Vector Regression 16.5 Implementation 16.6 Data Collection 16.7 Missing Value Processing 16.8 Feature Selection 16.9 Data Transformation and Feature Scaling 16.9.1 Error Metrics 16.10 Results and Discussion 16.11 Conclusion References Index