Big Data and Artificial Intelligence for Healthcare Applications

Cover
Half Title
Series Page
Title Page
Copyright Page
Table of Contents
Preface
Editors
Contributors
Part I Conceptual
Chapter 1 Introduction to Big Data
1.1 Big Data: Introduction
1.2 Big Data: 5 Vs
1.2.1 Volume
1.2.2 Velocity
1.2.3 Variety
1.2.4 Veracity
1.2.5 Value
1.3 Big Data: Types of Data
1.3.1 Structured Data
1.3.2 Semi structured Data
1.3.3 Unstructured Data
1.4 Big Data Analysis: Tools and Their Installations
1.5 Big Data: Commands
1.6 Big Data: Applications
1.7 Big Data: Challenges
References
Chapter 2 Introduction to Machine Learning
2.1 Introduction to Machine Learning
2.2 Artificial Intelligence
2.3 Python Libraries Used in Machine Learning
2.4 Classification of Machine Learning Based on Signals and Feedback
2.4.1 Supervised Learning
2.4.2 Unsupervised Learning
2.4.3 Reinforcement Learning
2.4.4 Semisupervised Learning
2.5 Data Preprocessing Using Python in Machine Learning
2.6 Types of Machine Learning on the Basis of Output to Be Predicted
2.6.1 Regression
2.6.2 Classification
2.6.3 Clustering
2.7 Natural Language Processing for Big Data
2.8 Big Data with Deep Learning
2.9 How Machine Learning Can Be Applied to Big Data
2.10 Machine Learning in Healthcare
References
Part II Application
Chapter 3 Machine Learning in Clinical Trials: A New Era
3.1 Introduction
3.2 ML-Based Algorithms and Methods
3.2.1 Support Vector Machine
3.2.2 Decision Trees
3.3 ML-Based Devices in Clinical Trials
3.3.1 ML-Based Classifiers and Sensor Data to Diagnose Neurological Issues in Stroke Patients
3.3.2 Detection of Severe Wounds and the Risk of Infection
3.3.3 Bone Age Analysis
3.3.4 Smart Watch and Shoulder Physiotherapy
3.4 Machine Learning in the Healthcare Sector
3.5 Machine Learning in Clinical Trials
3.5.1 Alzheimer’s Disease (AD)
3.5.2 Parkinson’s Disease (PD)
3.5.3 Attention-Deficit Hyperactivity Disorder (ADHD)
3.5.4 Cancer
3.5.5 Heart and the Circulatory System Disorders
3.6 Challenges and Future Scope in Clinical Trials with Machine Learning
3.6.1 Future Scope
3.7 Conclusion
References
Chapter 4 Deep Learning and Its Biological and Biomedical Applications
4.1 Introduction
4.1.1 (I) Application of Deep Learning in Biological Data Analysis
4.2 Read Quality Control Analysis
4.2.1 MiniScrub: De Novo Nanopore Read Quality Improvement Using Deep Learning
4.2.2 Deepbinner
4.3 Genome Assembly
4.3.1 CONNET: Accurate Genome Consensus in Assembling Nanopore Sequencing Data via Deep Learning
4.4 Application in Metagenomics
4.4.1 DeepARG: A Deep Learning Approach for Predicting Antibiotic Resistance Genes (ARGs) from Metagenomic Data
4.4.2 NanoARG: A Web Service for Detecting and Contextualizing Antimicrobial Resistance Genes from Nanopore-Derived Metagenomes
4.4.3 DeepBGC: A Deep Learning Genome-Mining Strategy for Biosynthetic Gene Cluster (BGC) Prediction
4.4.4 MetaPheno: A Critical Evaluation of Deep Learning and Machine Learning in Metagenome-Based Disease Prediction
4.4.5 MetagenomicDC
4.4.6 DEEPre: Sequence-Based Enzyme Commission (EC) Number Prediction by Deep Learning
4.4.7 DeepMAsED: Evaluating the Quality of Metagenomic Assemblies
4.4.8 DeepMicrobes: Taxonomic Classification for Metagenomics with Deep Learning
4.4.9 Meta-MFDL: Gene Prediction in Metagenomic Fragments with Deep Learning
4.4.10 IDMIL
4.5 Variant Calling from NGS Data
4.5.1 GARFIELD-NGS: Genomic vARiants FIltering by dEep Learning moDels in NGS
4.5.2 DeepSVR: A Deep Learning Approach to Automate Refinement of Somatic Variant Calling from Cancer Sequencing Data
4.5.3 DeepVariant: A Universal SNP and Small-Indel Variant Caller Using Deep Neural Networks
4.5.4 Clairvoyante: A Multi-Task Convolutional Deep Neural Network for Variant Calling in Single-Molecule Sequencing
4.6 SNP Effect Prediction
4.6.1 DeepSEA: Predicting Effects of Non-Coding Variants with Deep-Learning-Based Sequence Model
4.6.2 DANN: A Deep Learning Approach for Annotating the Pathogenicity of Genetic Variants
4.6.3 DeepMAsED
4.6.4 DeFine
4.7 Gene Expression Analysis (Bulk RNASEq, Single-Cell RNAseq)
4.7.1 Decode
4.7.2 DESC
4.7.3 scAnCluster: Integrating Deep-Supervised, Self-Supervised, and Unsupervised Learning for Single-Cell RNA-seq Clustering and Annotation
4.7.4 Digitaldlsorter: Deep-Learning on scRNA-Seq to Deconvolute Gene Expression Data
4.8 Transcription Factor/Enhancer (ChipSeq)
4.8.1 Enhancer Recognition and Prediction during Spermatogenesis Based on Deep Convolutional Neural Networks
4.8.2 DeepEnhancer: Predicting enhancers with deep convolutional neural networks
4.8.3 DeepHistone: A Deep Learning Approach to Predict Histone Modifications
4.8.4 An Integrative Framework for Combining Sequence and Epigenomic Data to Predict Transcription Factor Binding Sites (TFBSs) Using Deep Learning
4.9 RNA Protein Interaction Prediction
4.9.1 Recent Methodology Progress of Deep Learning for RNA-Protein Interaction Prediction
4.9.2 iDEEP: RNA-Protein Binding Motifs Mining with a New Hybrid Deep-Learning-Based Cross-Domain Knowledge Integration Approach
4.9.3 iDEEPS: Prediction of RNA-Protein Sequence and Structure Binding Preferences Using Deep Convolutional and Recurrent Neural Networks
4.9.4 Applications of Deep Learning in Biomedical Research
4.10 Deep Learning in Disease Diagnosis
4.10.1 Breast Cancer Screening
Commercially Available Solutions
4.10.2 Early Melanoma Detection
Commercially Available Solutions
4.10.3 Lung Cancer Screening
Commercially Available Solutions
4.10.4 Diabetic Retinopathy Screening
Commercially Available Solutions
4.10.5 Cardiac Risk Assessment from Electrocardiograms (ECGs)
Commercially Available Solution
4.10.6 Early Stroke Diagnosis from Head CT Scans
Commercially Available Solutions
4.11 Deep Learning in Diagnosis of Rare Diseases
4.12 Conclusions
Bibliography
Chapter 5 Applications of Machine Learning Algorithms to Cancer Data
5.1 Introduction
5.2 Overview of Feature Selection Approaches
5.2.1 Main Steps in Feature Selection
5.2.1.1 Preprocessing Step
5.2.1.2 Determining the Direction of Selection
5.2.1.3 Determining the Stopping Criteria
5.2.1.4 Evaluating the Selection
5.2.1.5 Validation Methods
5.2.2 Challenges in Feature Selection
5.3 Overview of Classification Methods
5.3.1 Overview of Popular ML Algorithms
5.4 Recent Applications of ML in Cancer Diagnosis/Classification
5.5 Web-Based Tools
5.6 Conclusion
References
Chapter 6 Pancreatic Cancer Detection by an Integrated Level Set-Based Deep Learning Model
6.1 Introduction
6.2 Related Work
6.3 Integrated Level Set-Based Deep Learning
6.4 Laplacian-Based Preprocessing
6.5 Integrated Level Set-Based Segmentation
6.6 N-Ternary Patterns
6.7 CNN-Based Deep Learning
6.8 Performance Analysis
6.9 Conclusion
References
Chapter 7 Early and Precision-Oriented Detection of Cervical Cancer: A Deep-Learning-Based Framework
7.1 Introduction
7.2 Deep Learning Networks and Cervical Cancer
7.3 Deep Learning Models
7.4 Deep-Learning-Based Classification of Cervical Cancer
7.4.1 Image Preprocessing and Data Augmentation
7.4.2 Region of Interest Extraction
7.4.3 Feature Extraction and Mapping
7.4.4 Cervical Cancer Classification
7.5 Results and Observations
7.5.1 Cervical Cancer Dataset
7.5.2 Model Validation
7.5.3 Classification Results
7.6 Limitations of Deep Learning for Cancer Prediction and Future Possibilities
7.7 Conclusion
Acknowledgments
References
Chapter 8 Transformation of mHealth in Society
8.1 What Is mHealth?
8.2 P’s of mHealth
8.3 Constituents of mHealth
8.3.1 The Sensory Device or Mobile Health
8.3.2 Telehealth
8.3.3 Electronic Health Records
8.3.4 Wireless Health
8.4 Services Offered by mHealth
8.5 Penetration of mHealth into Society
8.6 Distribution of Smart Devices
8.7 Reasons behind Success and Failures of mHealth
8.8 mHealth and Sister Technologies
8.9 Limitations and Regulations
8.9.1 Privacy Policy and Terms and Conditions
8.9.2 Request (Explicit) Consent
8.9.3 Making a Choice (Multiple Purposes)
8.9.4 Access to User’s Data
8.9.5 Privacy Dashboard
8.9.6 Permission Customization
8.9.7 The Right to be Forgotten
8.9.8 Sensitive Data
8.10 Promises and Challenges of mHealth
References
Chapter 9 Artificial Intelligence and Deep Learning for Medical Diagnosis and Treatment
9.1 Introduction
9.2 Deep Learning
9.3 Deep Learning Architecture
9.4 Types of Deep Learning Algorithms
9.5 Deep Learning Libraries
9.6 Application to Medical Diagnosis and Treatment
9.6.1 Imaging
9.6.2 Genomics
9.7 Tutorial
References
Part III Ethics
Chapter 10 Ethical Issues and Challenges with Artificial Intelligence in Healthcare
10.1 Medical Ethics
10.1.1 A History of Medical Ethics: From the Hippocratic Oath to the Nuremberg Code and Beyond
10.1.1.1 Definitions
10.1.1.2 The Hippocratic Oath and Its Values
10.1.1.3 From the Nuremberg Code to the Oviedo Convention
10.1.2 The Philosophical Foundations of Medical Ethics: Deontology and Teleology
10.1.2.1 Ethical Dilemmas in Medical Ethics
10.1.2.2 Deontological Ethics
10.1.2.3 Teleological ethics (or consequentialism)
10.1.3 From “Principlism” to Virtue Ethics:
10.1.3.1 The “Principlism” of Beauchamp and Childress
10.1.3.2 Virtue Ethics
10.1.3.3 Synthetic Tables of Ethical Theories (Table 10.2)
10.2 Ethics of Artificial Intelligence
10.2.1 Ethics of Technology and Ethics of AI
10.2.2 From the Myths of Artificial Intelligence to Neo-Luddism: A Plea for Regulation of AI
10.2.3 AI for Social Good: A New Principlism?
10.2.3.1 What Ethics for AI ?
10.2.3.2 Principles and Factors
10.2.3.3 Implementation
10.2.3.4 Tensions
10.2.3.5 Limitations
10.3 Principlism, Virtue Ethics, and AI in Medicine
10.3.1 A New Principlism for AI in Medicine
10.3.2 The New Principlism in a Clinical and Public Health Context
10.3.3 The New Principlism and Virtue Ethics: For a Responsible AI in Healthcare
Notes
References
Chapter 11 Epistemological Issues and Challenges with Artificial Intelligence in Healthcare
11.1 Key Issues in Philosophy of Artificial Intelligence (AI)
11.1.1 A Short History of AI: From 17th Century’s Calculators to the Dartmouth Artificial Intelligence Conference (2005)
11.1.1.1 From Pascal to Turing: A Prehistory of AI
11.1.1.2 From Turing to Dartmouth: A First Definition of AI
11.1.1.3 Neural Networks versus Expert Systems
11.1.1.4 Emergence and Development: Neural Networks and Machine Learning (ML)
11.1.2 The Modern Concepts of Artificial Intelligence and Big Data
11.1.3 AI: From Data to Knowledge
11.1.3.1 Truth and Proof
11.1.3.2 Machine Learning: Logic, Reliability, and Knowledge
11.1.3.3 From Proof to Trustworthiness
11.1.3.4 Explicability and Dependability
11.1.3.5 Biases and Errors
11.1.4 AI and Big Data: From the Myth to a New Paradigm?
11.2 Key Issues in Philosophy of Medicine: Medicine, Health, and Disease
11.2.1 The Hippocratic Conception of Medicine, Health, and Disease
11.2.1.1 Hippocrates and the Birth of Clinical Medicine
11.2.1.2 Hippocrates and the Theory of Four Humors
11.2.1.3 Hippocrates’ Conception of Health and Disease
11.2.2 The Modern Conception of Medicine, Health, and Disease: From Anatomo-Clinical to Experimental Medicine
11.2.2.1 Bichat and the Foundations of Anatomo-Clinical Medicine
11.2.2.2 C. Bernard and the Foundations of Experimental Medicine
11.2.2.3 The Modern Conception of Health and Disease
11.2.3 The Contemporary Conception of Medicine, Health, and Disease
11.2.3.1 The Molecularization of Medicine in the 20th Century
11.2.3.2 The Quantification of Medicine45 and the Birth of Evidence-Based Medicine (EBM)
11.2.3.3 From the Medicine of Disease to the Medicine of Health
11.3 Conclusion: Personalized Medicine or the Introduction of AI and Big Data in Medicine
Notes
References
Index