Machine Learning and Analytics in Healthcare Systems: Principles and Applications

Cover
Half Title
Series Page
Title Page
Copyright Page
Table of Contents
Preface
Editors
Contributors
Chapter 1 Data Analytics in Healthcare Systems – Principles, Challenges, and Applications
1.1 Introduction
1.1.1 Data Analytics in Healthcare
1.1.2 Characteristics of Big Data
1.2 Architectural Framework
1.2.1 Data Aggregation
1.2.2 Data Processing
1.2.3 Data Visualization
1.3 Data Analytics Tools in Healthcare
1.3.1 Data Integration Tools
1.3.2 Searching and Processing Tools
1.3.3 Machine Learning Tools
1.3.4 Real-Time and Streaming Data Processing Tools
1.3.5 Visual Data Analytical Tools
1.4 Data Analytics Techniques in Healthcare
1.5 Applications of Data Analytics in Healthcare
1.6 Challenges Associated with Healthcare Data
1.7 Conclusion
References
Chapter 2 Systematic View and Impact of Machine Learning in Healthcare Systems
2.1 Introduction
2.2 Applied ML in Health Care
2.2.1 ML-Assisted Radiology and Pathology
2.2.1.1 ML For Increased Imaging Precision in Radiology
2.2.2 Identification of Rare Diseases
2.2.2.1 Regular Challenges
2.2.3 ML in Mental Health Care
2.3 Major Applications
2.3.1 Personalized Medicine
2.3.1.1 Viable Personalized Medicine
2.3.2 Autonomous Robotic Surgery
2.4 ML in Cancer Diagnostics
2.4.1 ML and Cancer Imaging
2.4.1.1 Convolutional Neural Network (CNN) Imaging
2.4.1.2 Radiographic Imaging
2.4.1.3 Digital Pathology and Image Specimens
2.4.1.4 Image Database
2.4.2 Cancer Stage Prediction
2.4.2.1 Determination of Tumor Aggression Score (TAS)
2.4.2.2 AI Analysis of ML Models
2.4.3 Neural Network for Treatment Procedure
2.4.3.1 Classification and Prediction Modeling
2.4.3.2 Data Collection
2.4.4 Prediction of Cancer Susceptibility
2.5 Conclusion
References
Chapter 3 Foundation of Machine Learning-Based Data Classification Techniques for Health Care
3.1 Introduction
3.2 Machine Learning Techniques
3.3 Supervised Learning
3.4 Classification
3.4.1 Decision Tree
3.4.2 Random Forest
3.4.3 KNN Algorithm
3.4.4 Naïve Bayes
3.4.5 Neural Networks
3.4.5.1 Back Propagation in ANN
3.4.6 Support Vector Machines (SVM)
3.4.6.1 Advantages of SVM Classifiers
3.4.6.2 Disadvantages of SVM Classifiers
3.4.6.3 Applications of SVM
3.5 Regression
3.5.1 Logistic Regression
3.6 Unsupervised Learning
3.7 Clustering
3.7.1 K-Means Clustering
3.8 Applications of Machine Learning in Healthcare
3.8.1 Patterns of Imaging Analytics
3.8.2 Personalized Treatment
3.8.3 Discovery and Manufacturing of Drugs
3.8.4 Identifying Diseases and Diagnosis
3.8.5 Robotic Surgery
3.8.6 Clinical Trial Research
3.8.7 Predicting Epidemic Outbreaks
3.8.8 Improved Radiotherapy
3.8.9 Maintaining Healthcare Records
3.9 Conclusion
References
Chapter 4 Deep Learning for Computer-Aided Medical Diagnosis
4.1 Introduction
4.2 Computer-Aided Medical Diagnosis
4.2.1 Radiography
4.2.2 Magnetic Resonance Imaging (MRI)
4.2.3 Ultrasound
4.2.4 Thermography
4.2.5 Nuclear Medicine Imaging (NMI)
4.3 Deep Learning in Health Care
4.3.1 Deep Learning Neural Network
4.3.2 Imaging Analytics and Diagnostics
4.3.3 Natural Language Processing in Health Care
4.3.4 Drug Discovery and Precision Medicine
4.3.5 Clinical Decision Support Systems
4.4 Deep Learning vs CAMD
4.4.1 CAMD for Neurodegenerative Diseases
4.4.2 Deep Learning and Regularization Techniques
4.4.2.1 Multi-Task Learning
4.4.2.2 Convolutional Neural Network
4.4.2.3 Transfer Learning
4.4.3 CAMD and Big Medical Data
4.4.4 Deep Learning for Cancer Location
4.5 DL Applications in Health Care
4.5.1 Electronic Health Records
4.5.2 Drug Discovery
4.5.3 Medical Imaging
4.5.3.1 Image Analysis to Detect Tumors
4.5.3.2 Detecting Cancerous Cells
4.5.3.3 Detecting Osteoarthritis from an MRI Scan
4.5.4 Genome
4.5.5 Automatic Treatment
4.6 Major Challenges
4.6.1 Limited Dataset
4.6.2 Privacy and Legal Issues
4.6.3 Process Standardization
4.7 Conclusion
References
Chapter 5 Machine Learning Classifiers in Health Care
5.1 Introduction
5.1.1 Supervised learning
5.1.2 Unsupervised learning
5.1.3 Semi-supervised learning
5.1.4 Reinforcement learning
5.2 Decision Making in Health Care
5.2.1 Introduction
5.2.2 Clinical Decision Making
5.3 Machine Learning in Health Care
5.3.1 Introduction
5.3.2 Opportunities for ML in Health Care
5.4 Data Classification Techniques in Health Care
5.4.1 Support Vector Machine
5.4.2 Logistic Regression
5.4.3 Artificial Neural Network
5.4.4 Random Forest
5.4.5 Decision Tree
5.4.6 K-Nearest Neighbor
5.4.7 Naïve Bayes
5.5 Case Studies
5.5.1 Brain Tumor Classification
5.5.1.1 MRI Brain Image Acquisitions
5.5.1.2 Preprocessing
5.5.1.3 Convolutional Neural Network (CNN) Algorithm
5.5.1.4 Training of the Network
5.5.1.5 Validation of Data Set
5.5.1.6 Results
5.5.2 Breast Cancer Classification
5.5.3 Classification of Chronic Kidney Disease
5.5.4 Classification of COVID-19
5.6 Conclusion
References
Chapter 6 Machine Learning Approaches for Analysis in Healthcare Informatics
6.1 Introduction
6.1.1 Learning
6.2 Machine Learning
6.2.1 Types of Machine Learning
6.2.2 Different Algorithms
6.3 Supervised Learning
6.3.1 Regression
6.3.2 Classification
6.4 Unsupervised Learning
6.4.1 K-means Algorithm
6.4.2 Self-Organizing Feature Map (SOM)
6.5 Evolutionary Learning
6.5.1 Genetic Algorithm
6.5.1.1 Establishing the Genetic Algorithm
6.6 Reinforcement Learning
6.6.1 Markov Decision Process
6.7 Healthcare Informatics
6.7.1 Health Care
6.7.2 Applications
6.8 Analysis and Diagnosis
6.8.1 Analysis
6.8.2 Diagnosis
6.9 Machine Learning in Health Care
6.9.1 Overview
6.9.2 Types
6.9.3 Applications
6.9.4 Modules
6.10 Conclusion
References
Chapter 7 Prediction of Epidemic Disease Outbreaks, Using Machine Learning
7.1 Introduction
7.2 Predictive Analytics
7.2.1 Role of Predictive Analytics in Healthcare
7.3 Machine Learning
7.3.1 Machine Learning Process
7.3.1.1 Main Steps in Machine Learning Process
7.3.2 Types of Machine Learning Algorithms
7.3.2.1 Supervised Learning
7.3.2.2 Unsupervised Learning
7.3.2.3 Semi-Supervised Learning
7.3.2.4 Reinforcement Learning
7.4 Machine Learning Models for Predicting an Epidemic Disease Outbreak
7.4.1 Collection and Cleaning of Epidemic Disease Outbreak Data
7.4.1.1 Data Collection
7.4.1.2 Data Cleaning
7.4.2 Training the Model and Making Predictions, Using Machine Learning Predictive Analytics
7.4.2.1 Training the Model
7.4.2.2 Prediction
7.4.3 Results Visualization and Communication
7.5 Epidemic Disease Dissemination Factors
7.5.1 Physical Network
7.5.1.1 Population Density
7.5.1.2 Hotspots
7.5.2 Geographical Locations
7.5.2.1 Climatic Factors
7.5.2.2 Geodemographic Factors
7.5.3 Clinical Studies
7.5.3.1 Clinical Case Classification
7.5.3.2 Vaccination Tracking
7.5.4 Social Media
7.5.4.1 Geo-Mapping
7.6 Machine Learning Algorithms for Disease Epidemic Prediction
7.6.1 Support Vector Machine (SVM)
7.6.2 Decision Tree
7.6.3 Naïve Bayes
7.6.4 Artificial Neural Networks (ANNs)
7.6.5 K-Means Clustering
7.7 Existing Research on Machine Learning Application in Epidemic Prediction
7.8 Real-Time Epidemic Disease Prediction: Challenges and Opportunities
7.8.1 Challenges
7.8.2 Opportunities and Advances
7.9 Relevance of Machine Learning to the Novel Coronavirus (COVID-19) Outbreak
7.9.1 Design and Development of Vaccines and Drugs
7.9.2 Predicting the Spread of Virus, Using Social Media Platforms
7.9.3 Diagnosing Virus Infection via Medical Images
7.9.4 AI-Based Chatbots for Diagnosis
7.9.5 Smartphone Application Developments
References
Chapter 8 Machine Learning–Based Case Studies for Healthcare Analytics: Electronic Health Records, Smart Health Monitoring, Disease Prediction, Precision Medicine, and Clinical Support Systems
8.1 Introduction
8.2 Electronic Health Records
8.2.1 Supervised Machine Learning with EHR in Healthcare
8.2.2 Semi-Supervised Machine Learning with EHRs in Health Care
8.2.3 Unsupervised Machine Learning with EHR in Health Care
8.3 Smart Health Monitoring
8.4 Disease Prediction
8.4.1 Predicting the Presence of Heart Diseases
8.5 Precision Medicine
8.6 Clinical Decision Support System
8.6.1 Smart CDSS Architecture
8.7 Key Challenges
8.8 Conclusion and Future Directions
References
Chapter 9 Applications of Computational Methods and Modeling in Drug Delivery
9.1 Introduction
9.2 Computer-Aided Design for Formulation
9.2.1 Advantages of CADD
9.2.2 CADD Approaches
9.2.2.1 Structure-Based Drug Design (SBDD)
9.2.2.2 Ligand-Based Drug Design
9.3 Molecular Dynamics
9.4 Molecular Docking
9.4.1 Application of Docking
9.4.1.1 Hit Identification
9.4.1.2 Lead Optimization
9.4.1.3 Bioremediation Protein
9.5 Advances in Deep Learning Approaches
9.5.1 Artificial Neural Network
9.5.1.1 Preformulation
9.5.1.2 ANN for Structure Retention Relationship
9.5.1.3 Pharmaceutical Formulation Optimization
9.5.1.4 In-Vitro/In-Vivo Correlations
9.5.1.5 ANN in Quality Structure–Activity Relationships
9.5.1.6 ANN in Proteomics and Genomics
9.5.1.7 ANN in Pharmacokinetics
9.5.1.8 ANN in the Permeability of Skin and Blood Brain Barrier
9.5.1.9 Diagnosis of Disease
9.5.2 Convolutional Neural Networks (CNN)
9.6 Application of Computer-Aided Techniques to Pharmaceutical Emulsion Development
9.7 Application of Computer-Aided Techniques to the Microemulsion Drug Carrier Development
9.8 Applications of Multiscale Methods in Drug Discovery
9.8.1 Approaches of Multiscale Modeling
9.8.1.1 Cardiac Modeling Molecular Dynamics
9.8.1.2 Network Biology and Cancer Modeling
9.9 Accelerated Drug Development by Machine Learning Methods
9.10 Conclusion
References
Chapter 10 Healthcare Data Analytics Using Business Intelligence Tool
10.1 Introduction: Big Data
10.2 Data Collection
10.2.1 Electronic Health Records (EHR)
10.2.2 Laboratory and Diagnostic Reports
10.2.3 Prescriptions
10.2.4 Forms Filled By Patients
10.3 Data Pre-Processing
10.3.1 Data Selection
10.3.2 Data Cleansing
10.3.3 Data Conversion/Transformation
10.3.4 Data Integration
10.4 Data Analytics and BI
10.4.1 Data Source Identification
10.4.2 Data Staging
10.4.3 Data Warehouse/Mart
10.4.4 Data Analysis
10.4.5 Visualization and Reporting
10.4.6 Decision Making
10.5 Business Intelligence Tools
10.5.1 Introduction to Power-BI
10.5.2 Results and Discussion
10.5.2.1 Getting Started with Power-BI
10.5.2.2 Working with Multiple Data Sources
10.5.2.3 Creation and Sharing of Dashboard
10.5.3 Recommender System
10.6 Findings from EHR Records, Using Machine Learning Algorithms
10.6.1 Descriptive Analytics
10.6.2 Predictive Analytics and Insights
10.7 Conclusion
References
Chapter 11 Machine Learning-Based Data Classification Techniques in Healthcare Using Massive Online Analysis Framework
11.1 Introduction
11.1.1 Disease Identification and Diagnosis
11.1.2 Drug Discovery
11.1.3 Medical Imaging
11.1.4 Personalized Medicine
11.1.5 Smart Health Records
11.1.6 Disease Prediction
11.2 Types of Healthcare Data
11.2.1 Clinical Data
11.2.2 Sensor Data
11.2.3 Omics Data
11.3 Time-Series Data in Healthcare
11.3.1 Time-Series Analysis
11.4 Machine Learning Algorithms on Classification Tasks
11.4.1 Classification
11.4.2 Classification Model
11.4.3 Binary Classification
11.4.4 Multi-Class Classification
11.4.5 Multi-Label Classification
11.4.6 Imbalanced Classification
11.5 Massive Online Analysis (MOA) Framework for Time-Series Data
11.5.1 Setting the Environment
11.5.1.1 Upload Dataset or Generating Synthetic Dataset
11.5.1.2 Converting CSV File to .arff File Format
11.5.2 Data Generator
11.5.3 Learning Algorithms
11.5.4 Evaluation Methods
11.5.4.1 Holdout Estimation Method
11.5.4.2 Prequential or Interleaved Test-Then-Train Estimation Method
11.5.4.3 Evaluation Performance Metrics
11.5.5 Discussions on Results
11.6 Laboratory Exercise and Solutions
11.7 Conclusion
References
Chapter 12 Prediction of Coronavirus (COVID-19) Disease Health Monitoring with Clinical Support System and Its Objectives
12.1 Introduction
12.2 History of COVID-19
12.2.1 Coronavirus
12.2.2 Global Health Security
12.2.3 Types of Coronavirus in Human Beings
12.2.3.1 General Coronavirus in Human Beings
12.2.3.2 Other Human Coronaviruses
12.2.3.3 SARS-CoV (Severe Acute Respiratory Syndrome Coronavirus)
12.2.3.4 MERS-CoV (Middle East Respiratory Syndrome Coronavirus)
12.2.3.5 SARS-CoV-2 (Severe Acute Respiratory Syndrome Coronavirus 2)
12.3 Inter-Relations between Artificial Intelligence, Machine Learning, and Deep Learning
12.3.1 Machine Learning
12.3.1.1 Problem Types Solved through Machine Learning
12.3.1.2 Types of Machine Learning Algorithms
12.3.2 Machine Learning Workflow
12.3.2.1 Smart Health Monitoring System
12.3.2.2 Electronic Health Records (Electronic Medical Records)
12.3.2.3 Manipulation of Supervised Concern and the Incorporated Delivery Scheme
12.3.2.4 Functional Operation of an Electronic Health Record System
12.3.2.5 Inquiry and Inspection Systems
12.3.2.6 Medical Care
12.3.2.7 Experimental Study
12.4 Conclusion
References
Index