Foreword to Artificial Intelligence in Medicine
Quotation
Preface
Acknowledgments
In Memoriam
Contents
About the Editors
Contributors
Part I:
1 Basic Concepts of Artificial Intelligence: Primed for Clinicians
Introduction
AI, Machine Learning, and Deep Learning per Definition
A Brief History of AI
Rising Demand for AI
AI Applications
AI Staging
AI Programming Languages
Machine Learning
Types of Machine Learning
Machine Learning Problem Solutions
Supervised Learning Algorithms
Unsupervised Learning Algorithms
Reinforcement Learning
Limitations of Machine Learning
Introduction of Deep Learning
Single-Layer Perceptrons
Multilayer Perceptron - Artificial Neural Network
Natural Language Processing
Conclusions
References
2 Applying Principles from Medicine Back to Artificial Intelligence
Introduction
Concepts from the Medical Sciences Being Applied Back to the Field of Artificial Intelligence: Neural Networks
Concepts from the Medical Sciences Being Applied Back to the Field of Artificial Intelligence: Cognitive Architectures
Concepts from the Medical Sciences Being Applied Back to the Field of Artificial Intelligence: The Causal Cognitive Architectu...
Discussion
References
3 Mathematical Foundations of AIM
Introduction
Classical Machine Learning and Its Limitations
The Perceptron Model
Support Vector Machine
Modern Deep Learning Revolution
Architectures of Modern Deep Neural Networks
Representation Power of Deep Neural Networks
Other Properties of Deep Neural Networks
Algorithm Unrolling: From Iterative Algorithms to Deep Networks
Learned Iterative Shrinkage and Thresholding Algorithm
Unrolling Generic Iterative Algorithms
Interpretations of Deep Learning
Hierarchical Features in the Visual System
Geometric Understanding of Deep Neural Networks
Summary and Outlook
References
Part II:
4 Introductory Approaches for Applying Artificial Intelligence in Clinical Medicine
Introduction
Short AIM History
Electronic Health Records (EHRs) and AIM Interaction
Learning Health Systems
Some Industrial Cases - An Overview
AIM Applications
Financial Aspects of AIM
Emerging Markets
Clinical Decision Support Systems
Conclusion
References
5 Introduction to Artificial Intelligence in Medicine
Introduction
The Development of the AI Framework
How Does a Convolutional Neural Network Work?
Convolution and Cross-Correlation
A Close Look into ``AlexNet´´
Forward Pass
Backward Pass
Representation Learning
Unsupervised Learning and a Geometric Model
Network Topologies, Types of Learning and Performance Measures
Topologies of Networks
Types of Learning
Measures of Performance: Sensitivity, Specificity, ROC
Inference and Network Examples
Deep Neural Network Application Domains
Relation to the Visual System
Color Processing and Colorization
Foveated Vision
Discussion
Lessons for All Doctors
References
6 Importance of AI in Medicine
Introduction
The Amazing World of AI
A World of Terminology
Neural Nets: Quite Similar to Neural Networks in the Human Brain
Data: The Fuel for AI
Segmentation: Feeding Quality Data
Training Mode
Multisensor Data
Prediction Mode
Levels of Expertise
Self-Evaluation Mode and Correlations
Breaking Boundary Conditions
Multiple Ways of Solving a Problem
Processing Improvements
Validation
Gatekeepers
Limitations
Responsible Artificial Intelligence
Various Agendas
Transparency
When to Use AI in Medicine
References
7 The New Frontiers of AI in Medicine
Introduction
Natural Language Processing
Can Future AI Systems Use Natural Language Processing to Improve How Individuals Engage in Their Health?
Unlocking Clinical Knowledge to Address Common Patient Concerns Through Clinical Avatars
Engineering the Next-Generation Electronic Health Record and Improving Staff Workflows
Machine Learning, Deep Learning, and Neural Networks
Improved Disease Detection Through Computationally Enabled Diagnostics
Entirely New Ways of Diagnosing Disease Using Digital Biomarkers
AI-Driven Population-Level Interventions for Entire System Planning and Optimization
Computer Vision
Using AI to Generate Real-Time Synthetic Images to Improve Surgical Interventions
AI-Driven Surgical Success Measures to Create Intelligent System-Wide Resource Planning
Robotics
Partly Automated Surgery for Rapid Recovery and Safety
New Frontiers in AI
AI Bias Gives Another Perspective on Driving Improvement
Healthcare Service Redesign
Evidence-Based Individualized Treatment Pathways
The Future of AI in Medicine
References
8 Social and Legal Considerations for Artificial Intelligence in Medicine
Introduction
Social Challenges
Juristic Challenges
Tort Law
Conclusions and Guidelines
References
9 Ethical Challenges of Integrating AI into Healthcare
Respect for Autonomy
Beneficence
Nonmaleficence
Privacy
Safety
Justice
Impact on the Physician-Patient Relationship
References
10 Artificial Intelligence in Medicine and Privacy Preservation
Introduction
General Considerations and Current Technical Standards
Anonymization, Pseudonymization, and k-Anonymity
Considerations for Specific Dataset Types
The Requirement for Next-Generation Privacy-Preserving Techniques
Federated Learning
Technical Framework
Challenges in Federated Learning
Attacks Against Federated Learning Systems
Applications of Federated Learning
Differential Privacy
Properties
Implementation
Sensitivity and Privacy Budget
Challenges
Applications
Homomorphic Encryption and Secure Multi-Party Computation
Homomorphic Encryption
Applications
Secure Multi-Party Computation
Applications
Trusted Execution Environments
Outlook
References
11 Artificial Intelligence for Medical Decisions
Introduction
Automation of Decision-Making in Healthcare
Taxonomy of Medical Decisions
Logic-Based Methods
The Language of Logic
Knowledge Representation and Reasoning
Beyond First-Order Logic
Learning from Data
Statistical Modelling and Machine Learning
Three Machine-Learning Approaches
The Impact of the Deep Learning Revolution
Combinatorial Optimization Methods
Reinforcement Learning for Sequential Decision-Making
Bayesian Models for Decision Support
Bayesian Networks for CDSS
The Need for Causality in Clinical Decision-Making
Explainability, Interpretability, and Fairness
Conclusion
References
12 Artificial Intelligence for Medical Diagnosis
Introduction
Diagnostic Reasoning
Knowledge-Based Diagnosis
Rule-Based Diagnosis
Fuzzy-Logic Systems
Ontology-Based Systems
Model-Based Diagnosis
Abductive Diagnosis
Bayesian Diagnosis
Causal Reasoning for Diagnosis
Machine Learning for Diagnosis
Learning from Data
Machine Learning as Function Approximation
Three Supervised Methods
Other Machine Learning Formalisms for Diagnosis
The Importance of Data
Outlook
References
13 AIM and the History of Medicine
Introduction to the History of Cognitive Science and Intelligence
Mechanical and Biological Automatons
The Golem
The Ars Magna
The Concept of Symbolic Languages
Hurufism
The Calculator
The Technological Myth Persists
AI is on its Way
Conceptual Revolutions
World, Meet the Personal Computer
The Big Question: Can Computers Think?
Earliest Steps towards Computerized Medicine
Collaboration Between Medical Scientists and AI Researchers
Examples of the Use of AI in Medicine
Artificial and Biological Life
Conclusion
References
14 AIM and Patient Safety
Introduction
Trends in Quality and Safety Research
Approaches to Patient Safety and Preventing Errors
Intelligent Systems: Machine Learning and Natural Language Processing
Prevention of Adverse Events
Diagnostics
Medication Errors and Polypharmacy
Treatment Outcomes and Quality
Patient Safety Databases
Future of AI in Safety and Quality
References
15 Right to Contest AI Diagnostics
Introduction
The Right to Effective Contestability and Transparency
The Four Dimensions of Contestability
Dimension 1: Personal Health Data
Dimension 2: Bias
Dimension 3: Performance
Dimension 4: Decisional Role
Further Issues and Aspects of the Right to Contest
References
16 AIM in Medical Informatics
Introduction
Clinical Data
Electronic Health Record
Omics Data
Diagnostic-Related Information and Treatment Information
Data Processing
Missing Value Imputation
Dimensionality Reduction
Different Omics Processing
Existing and Emerging Applications in Medical Diagnosis
Clinical Data Application
Omics Data Application
Explainability of Results
Future Perspectives
Conclusion
References
17 Artificial Intelligence in Evidence-Based Medicine
Introduction
From PICO Questions to Systematic Reviews
Automation of Systematic Reviews
Development of Search Strategies
Screening
Data Extraction
New Types of Evidence
Making Data More FAIR
Other Sources of Data
Improving Shared Decision-Making
Summary
Cross-References
References
18 AIM in Electronic Health Records (EHRs)
Introduction
The Symbolic-Connectionism Cognitive Architecture
The Holistic Framework of Research Work of WI Lab in AIM
Data Tier
Knowledge Tier
Application Tier
Disease Diagnosis
EMR Quality Control
Overtesting Detection
Disease Risk Prediction
Chronic Disease Management
Dermatosis Diagnosis
Selected Research
Corpus Construction
Information Extraction
Knowledge Expansion
Integrating External Knowledge
Mining Potential Knowledge
Clinical Reasoning Model
Knowledge Representation
Disease Diagnosis
Disease Risk Prediction
Conclusion and Future Work
References
19 AIM and Causality for Precision and Value-Based Healthcare
Current Standard of Care and Medical Research
Current AI Practice in Medicine and Healthcare
Model-Driven AI and Causal Diagnostics
The Opportunity for Precision Medicine
Conclusions
References
20 AIM and the Nexus of Security and Technology
Introduction
Human Perfection and the Security Implications of Biomedical Technology
Eugenics: The Painful History of an Idea
Can ``Liberal Eugenics´´ Really Work? The Relationship Between New Wave Eugenics and Security
Exploring the Symbiotic Relationship Between Technology and Security
Hacking Healthcare: The Violation of Patient Data
Concluding Thoughts
References
Bibliography
21 AIM in Unsupervised Data Mining
Introduction
Association Rule Mining
FRM
BRM
Likelihood Mining Criterion (LMC)
LMC-FRM Comparison
Basic Rule Mining Example
Methodology
Evaluation
Census and Chemical Exposure Database Mining
Methodology
Evaluation
Rehabilitation Routine Mining
Methodology
Evaluation
Conclusions
References
22 AIM in Medical Education
Introduction
AI in Various Specialty Delivered Medical Education
Literature Review
Inclusion and Exclusion Criteria
Meta-Analysis
Bias Assessment
Results
Results from Screening
Bias Assessment Result
Meta-Analysis
Discussion
Meta-Analysis
Proposing a New Method of Unbiased Reporting of Meta-Analysis of Improvement in AI Studies
The State of Medical and Surgical Training and the Use of AI in Simulation
AI in Medical Specialties for Education
AI in Ophthalmology, Dermatology, Cardiology, Gastroenterology, and Rheumatology
AI in Obstetrics and Gynecology, General Surgery, Orthopedic, and Neurosurgery Education
AI for Surgical Performance Assessment
AI for Precision Examination and Ethico-Legal Aspects of Medical Education and Autonomous Robotic Surgical Regulation
Conclusion
References
23 AIM and Evolutionary Theory
Introduction
Mathematical Oncology
Prediction of Cancer Risk
Precision Medicine
Future Directions
Infectious Disease
Drug Resistance
Drug Design
Emerging Pathogens
Avoiding the Desynthesis
References
24 AIM and the Patient´s Perspective
Introduction
AIM and the Patient´s Perspective
The Need for Transparency
Regulations and Data Sharing
The Public´s Understanding of AI
The Public´s View on Data Sharing
Weighing the Benefits and Risks
Levelling the Playing Field
Gaining the Public´s Trust
Technologies for Trust
Privacy by Design
Blockchain
Federated Computing
Dynamic Consent
Conclusion
References
25 AIM and Hackathon Events
Introduction
Organizing AIM Hackathons
Proposed Metrics for Measuring the Success of an AIM Hackathon
Limitations of the Hackathon Approach
Recommendations for Organizing Better Hackathons
References
Further Reading
26 AIM, Philosophy, and Ethics
Introduction
Promises of AI in Medicine
AI and Medical Epistemology: A Changing Paradigm
Data
Data-Utopianism
Data Curation and Use
AI and Medical Epistemology: Limits, Risks, and Biases
Human Biases and Prejudices: Language and Interpretation
Computational Biases: Programming and Algorithms
AI and Medical Ethics
The Patient-Doctor Relationship
The Medical Profession in the Era of Digital Capitalism
Conclusion and Recommendations
References
27 Reporting Standards and Quality Assessment Tools in Artificial Intelligence–Centered Healthcare Research
Introduction
The Case for AI-Specific Instruments
Specific AI Reporting Standards
SPIRIT-AI and CONSORT-AI
STARD-AI
TRIPOD-AI
Specific AI Quality Assessment Tools
QUADAS-AI
PROBAST-AI
Conclusion
References
28 AIM and Gender Aspects
Introduction
Sex and Gender Differences in Medicine
Sex and Gender Bias in Machine Learning Models
Role of Sex and Gender in Machine Learning Models for Medicine
Current Issues
Outlook and Potential Solutions
Collecting Balanced Datasets
Disaggregating Data
Data and Model Documentation
Fairness Aware Models
External Auditing Framework for Machine Learning Models
Continuous Monitoring of the Models
References
29 Meta Learning and the AI Learning Process
Introduction
General Considerations
Transfer Learning
Few-Shot Learning
Continual Learning
Multi-task Learning
Neural Architecture Search
Conclusion
References
30 Artificial Intelligence in Medicine Using Quantum Computing in the Future of Healthcare
Introduction
History
The Basics of Quantum Computingand Quantum Machine Learning
Types of Quantum Computers
Companies
Basic Anatomy of a Quantum Versus Classical Computing Circuit
The Bit and the Classical Logic Gate
The Qubit and the Quantum Gate [58]
State Vectors
The Quantum Gates [58]
Quantum Superposition and Machine Learning Applications
Theoretical Medical Applications
Quantum Tunnelling and Machine Learning Applications
Quantum Entanglement and MachineLearning Applications for Medicine and Surgery
Basic Mathematical Formalism of Entanglement Applied to Medicine [58, 74]
Important Quantum Computing Algorithms and Quantum Machine Learning Algorithms
Quantum Fourier Transform
Shor´s Algorithm
Grover´s Search Algorithm
Deutsch-Jozsa Algorithm
Bernstein-Vazirani Algorithm
Quantum Hidden Markov Chain Algorithm
Quantum Natural Language Processing
Quantum Annealing and Quantum Neural Networks
Quantum Enhanced Reinforcement Learning
Quantum Phenomenon in Disease
Quantum Computing in Healthcare
Future Translational Considerations for E-Healthcare
Ethico-Legal Implications and Considerations
Summary Remarks
References
Part III:
31 Emergence of Deep Machine Learning in Medicine
Introduction
Deep Neural Networks
The Universal Approximation Theorem and Its Limitation
Internal Hierarchical Feature Extraction
Internal Transformation of Dataset Topologies
Medical Examples
Medical Imaging
Genomics and Epigenomics
Natural Language Processing
Conclusion
References
32 AIM in Interventional Radiology
Introduction to Interventional Radiology
AI in Medical Imaging: A Primer
Data in Medical Imaging
AI in Diagnostic Radiology: Brief Overview
AI in Interventional Radiology: Unique Challenges
AI in Interventional Radiology: Opportunities
The Ideal Interventional Radiology Suite
Scope of AI in Interventional Radiology
AI in IR: Decision Support
AI in IR: Triaging of Patients
AI in IR: Prevention of Errors
AI in IR: Periprocedural Support
AI in IR: Patient Monitoring and Procedural Support
AI in IR: Prognostication and Outcome Prediction
AI in IR: Image Acquisition and Processing
AI in IR: Residency and Fellowship Training
Can AI Replace Diagnostic or Interventional Radiologists?
References
33 Automated Deep Learning for Medical Imaging
Introduction
Challenges of Deep Learning for Clinical Researchers
Highly Specialized Technical Expertise
Compute Resources
Large, Well-Curated Datasets
Data Protection and Privacy
Principles
Initial Work Utilizing Automated Deep Learning in Medicine
Automated Deep Learning Process
Limitations of Automated Deep Learning
General Limitations
Explainability
Generalizability and Bias
Automated Deep Learning-Specific Limitations
Future Directions of Automated Deep Learning
Use Cases of Automated Deep Learning
Packaging and Deployment of These Models
Conclusion
References
34 AI in Musculoskeletal Radiology
Introduction
Machine/Deep Learning
Musculoskeletal (MSK) Diseases: A Rapidly Rising Global Burden
Bringing Disease Management into the Digital Age
Musculoskeletal Radiology
AI in Computed Radiographs
Building and Validating AI Models for MSK Radiographs
Quality Control
Anomaly Detection
IB Lab KOALA: Assessment of the Gonarthrosis Stage
IB Lab PANDA: Determination of Pediatric Bone Age
IB Lab HIPPO: Measurement of the Hip and Pelvis
IB Lab LAMA: Measurement of the Whole Leg
AI Disrupts Status Quo of Radiology Reporting
Take-Home Message: Artificial Intelligence (AI) in Musculoskeletal Radiology
References
35 AIM and Explainable Methods in Medical Imaging and Diagnostics
Introduction
Medical Imaging in Clinical Diagnosis
Recent Advances in Technology Toward Medical Imaging
Machine Learning-Based Methods
Explainable Artificial Intelligence: XAI
Transparent Models
Post Hoc Interpretability
Explainable Models for Deep Neural Networks
Conclusions and Future Landscape for ML-Based Imaging and Diagnostic Methods
References
36 Optimizing Radiologic Detection of COVID-19
Introduction
Test Set Technologies
Artificial Intelligence in Medical Imaging and the Opportunity Provided by Test Sets
Why Does Education Need to Embrace AI to Enhance Detection of COVID-19?
AI-Tailored Education Using Clinician Demographics and Image Features
AI Companions to Help with COVID-19 Diagnosis
Finally, Some Cautionary Notes
Conclusion
References
37 AIM in Surgical Pathology
Introduction
Definitions
A Brief History
Clinical Applications
Detecting and Classifying Disease
Grading and Scoring Disease
Finding or Outlining Tumors and Tissue
Finding Rare Events and Small Objects
Predictive Tasks
Image Quality Tools
Grand Challenges
Technical Aspects of Digital Pathology
Standards of Reporting of AI
Deployment and Regulation
Technical
Infrastructure
People
Education
Conclusion
Cross-References
References
38 Artificial Intelligence in Kidney Pathology
Introduction
Detection
Detection Using Conventional Features
Detection Using Deep Learning
Segmentation
Segmentation of Glomeruli
Segmentation of Multiple Structures
Classification
Classification of Major Pathological Findings
Classification and Identification of Specific Components
Classification Based on Pathological Category
Classification of Images with Immunohistochemistry
Classification Based on the Clinical Category and Genotype
Summary and Future Implications
Equations
References
39 AIM in Dermatology
Introduction
AIM in Dermatology
Area of Application: Skin Cancer
Area of Application: Psoriasis
Area of Application: Eczema
Area of Application: Other Skin Disorders
Dermatologist Attitude Toward Artificial Intelligence
Limitation of Artificial Intelligence in Dermatology
Better Applicability of AI Thanks to Teledermatology
Ethnic Variations as a Challenge in the Development of Algorithms
References
40 Artificial Intelligence in Predicting Kidney Function and Acute Kidney Injury
Introduction
AKI Overview
Background for AKI Prediction
ML Model for AKI Onset Prediction
Definition of AKI Event
Prediction Timepoint and Target Period
Input Features
ML Algorithms
Model Performance
The External Validity of the Models
Explainability of Models
Implementation Challenges
Conclusion
References
41 AIM in Hemodialysis
Introduction
Anemia, Hemoglobin Prediction, and ESA Dosage Optimization
MPC-Based Approaches: Hemoglobin Prediction
Physiologically Based Models
AI-Based Models
Linear Regression
Feed-Forward Neural Networks
Recurrent Neural Networks
Other Supervised Learning Models
Direct Dose Optimization
Rule-Based Systems
Reinforcement Learning
Comorbidities, Mortality Prediction, and Patient Clustering
Other Miscellaneous Applications
Future Developments and Conclusions
Cross-References
References
42 Artificial Intelligence in Public Health
Introduction
Defining Public Health and Social Medicine: Toward Greater Precision, or a Regression from the Collective to the Individual?
Public Health Versus Individual Health, Social Medicine Versus Personalized Medicine
Precision Versus Individualization. Precision Public Health
Public Health, a Practice Based on Data, on Information, or on Evidence?
Where Is Artificial Intelligence Used in Public Health?
There Is No Artificial Intelligence Without Data: Data Federation and ``New´´ Types of Data
The Citizen and the Patient: Producer, Actor, and Manager of Their Own Health
AI and Health Surveillance Systems
Learning Healthcare Systems (LHSs)
Risk and Insurance
The Notion of Risk in the Era of AI
Approaches Adopted from Marketing and PR: Segmentation and Targeting
Organization and Governance
Capturing Data for Research and Governance
New Public Health Actors: The Role of Platforms and the Private Sector
Future Challenges
Challenges Shared with Other Health Fields
More Specific Challenges
Conclusion and Outlook
Cross-References
References
43 AIM and Business Models of Healthcare
Introduction
The Business Perspective: Product Development and Sales
The Consumer Perspective: Purchaser, End User, and Patient
Myth of Generalizability
Proposed Consideration 1: Co-creation
Proposed Consideration 2: Multi-Stakeholder Engagement
Proposed Consideration 3: Metrics for Defining Value Delivered
Ethics, Law, and Policy
Conclusion
References
44 AIM for Healthcare in Africa
Introduction
Brief History of Artificial Intelligence for Medicine in Africa
Current Applications of AI for Medicine in Africa
Challenges with AIM Landscape in Africa
Digital Health Foundation
Data Availability and Quality
Infrastructure
Costs and Funding
Governance, Regulations, and Ethics
Critical Areas of Attention for Improving AIM in Africa
Governance and Ethical Approaches
Building for the Africa (How Best to Approach Solutions and Implementation)
Opportunities for AIM Impact in Africa
Overview of Selected Applications of AI for Healthcare in Africa
Kenya Medical Supplies Agency (KEMSA) with IBM´s Watson
Afya Pap in Zimbabwe
Delft Institute´s CAD4TB Software
References
45 Aim in Climate Change and City Pollution
Introduction
Machine-Learning Methods in the Study of Urban Pollution
ML Methods in Air-Pollutant Modeling
ML Methods to Model Flow Dynamics
Remote Sensing for Urban Air Observation
Impact of Remote-Sensing Sensors for Monitoring Urban Airflow
Remote-Sensing Data Resources and Analysis Methods
Further Supportive Data that Satellite Remote Sensing Can Offer
Challenges and Open Problems
References
46 AIM in Pharmacology and Drug Discovery
Introduction
Ligand Screening and Pharmacology
Ligand-Based Approach
Structure-Based Approach
Chemical Genomics-Based Approach and Polypharmacology
ADME in Pharmacokinetics
Absorption
Distribution
Metabolism
Excretion
Data Source for Drug Discovery
Omics Data
Real-World Data
Summary and Future Implications
References
47 Clinical Evaluation of AI in Medicine
Clinical AI Systems Require Robust Clinical Evaluation
Randomized Controlled Trials of Clinical AI Systems
AI Systems for Disease Diagnosis
AI Systems for Detection of Colonic Adenoma During Endoscopy
An AI System for Detecting Blind Spots During Esophagogastroduodenoscopy
An AI System for Detecting Paroxysmal Atrial Fibrillation
An AI System for Diagnosing Childhood Cataracts
AI Systems for Disease Prediction
An AI Early Warning System for Detecting Intraoperative Hypotension
An AI System for Mental Health Risk Assessment
AI Systems for Adjusting Therapeutic Treatment
An AI System for Optimizing Insulin Dose
An AI System for Monitoring Drug Adherence
Reporting Standards for AI Clinical Trials
New Reporting Guidelines for AI Trials to Reflect the New Epoch
Future Challenges
References
48 Artificial Intelligence in Medicine: Biochemical 3D Modeling and Drug Discovery
Introduction
Predicting the 3D Structures of Proteins
Protein Folding
Secondary Structure Prediction
Ab Initio Tertiary Structure Prediction
AlphaFold
In Silico Drug Discovery
Drug Repurposing
Deep Neural Networks
Generative Adversarial Networks
Inputting Molecular Structures
Conclusion
References
49 AIM in Endocrinology
Introduction
AI Applications in Diabetes Mellitus
AI-Driven Diabetes Care: Changing the Landscape
Prediction of Diabetes Risk
Retinopathy Detection
Prediction of Diabetic Complications
Continuous Glucose Monitoring and Closed-Loop Artificial Pancreas System
Therapeutic Lifestyle Modification
AI Applications in Bone and Mineral Disorders
Fracture Identification
Opportunistic Screening of Osteoporosis and Sarcopenia
Fracture Risk Assessment
Finding Novel Biomarkers Related to Bone Metabolism
AI Applications in Thyroid Disorders
AI Application in Thyroid Cancer
AI Application in Functional Thyroid Disorders
AI Applications in Pituitary and Adrenal Disorders
Diagnosis and Subtyping
Prediction of Treatment Outcomes
Implications of AI for the Endocrinologists
References
50 Artificial Intelligence and Hypertension Management
Introduction
Artificial Intelligence Approaches for Hypertension Management
AI-Surrogate Measurement for BP
Surrogate BP Measurement Using Wearable Sensors
Surrogate BP Measurement Using Smartphone Cameras
Challenges
AI-Factor Analysis for BP Changes
Causal Inference Using ML for Hypertension
Explainable AI for Hypertension
Challenges
AI-Forecasting for Future BP
BP Forecasting with Cross-Sectional Data
BP Forecasting with Time-Series Data
Challenges
Conclusions
References
51 Aim and Diabetes
Introduction
Problems Faced by People with Diabetes
The AI Approach
Expert Systems Versus Decision Support Systems
Prevention and Prognosis
Diagnosis of Diabetes
Diabetes Management
Lifestyle Interventions/Behavioral Change
Risk Stratification
Complications and Comorbidities
Discussion
Conclusions
References
52 AIM in Primary Healthcare
Introduction
Opportunities of AI in Primary Care Include:
Shift of Balance in Healthcare
Electronic Health Records and Data Ownership
Global Macrotrends [1-3]
Symptom Checkers and Dissemination of Specialities
Altered Roles
Precision Medicine and Frontiers for AI
Legal and Regulatory Aspects
Privacy Concerns
Patient Safety
Medical Imaging Diagnostics and Radiology
Medical Informatics and Clinical Decision Support
Patient´s Perspective
Gender Aspects
Point-of-Care Dermatology and Ophthalmology
Public Health Aspects on Primary Healthcare
AI for General Practice Management
Endocrinology and Diabetics
Cardiovascular and Respiratory Management
Cardiac
Hypertension
Respiratory
Chronic Neurological and Neuropsychiatric Disease Monitoring
Obstetrics, Pregnancy, and Pediatrics
Oncology
Conclusion
Cross-References
References
53 AIM in Nursing Practice
Chapter Introduction
Introducing the IRs
Brief History of AI
Made not Born: The Beginning
Definition
Subsets of AI
Supervised Machine Learning (SML)
Unsupervised Machine Learning (UML)
Reinforcement Machine Learning (RML)
A Case for ICU
Deep Machine Learning (DML)
Nursing
Nursing Practice
Societal Change
AI and Nursing Education
Enhancing Holistic Care
Robots to the Rescue
The Future of Robotics
Conclusion
References
54 AIM in Respiratory Disorders
Introduction
Imaging
Chest X-ray
Computed Tomography (CT)
Histopathology
Bronchoscopy
Point-of-Care Ultrasound (PoCUS)
Lung Function Testing
Pulmonary Function Tests
Spirometry
Forced Oscillation Tests
Telemedicine
Miscellaneous
Sleep Monitoring
Breath Analysis
Lung Sound Analysis
Conclusions and Future Perspectives
References
55 AIM in Rheumatology
Introduction
Application of Artificial Intelligence in Rheumatology
Application in Electronic Health Records
Application in Genetic and Biomarker Data
Application in Medical Images
Application in Mixture Data
Future Perspectives and Challenges
Conclusions
References
56 AIM in Osteoporosis
Introduction
Methods
Nonsparse Classification Techniques: Texture-Based, Patch-Based, and Deep Learning
Classifiers and Discriminant Functions
Bag of Keypoints
Deep Neural Networks
Sparse Representation and Classification
Integrative Ensemble Sparse Analysis Techniques
Results
Discussion
References
57 Artificial Intelligence in Laboratory Medicine
Introduction
A Gentle Introduction: What Is Machine Learning?
A Brief Overview of Machine Learning Implementation in Laboratory Medicine
Machine Learning Models in Laboratory Medicine
Conclusion
References
58 Artificial Intelligence in Medicine (AIM) in Cardiovascular Disorders
Introduction
Overview of AI Research on Cardiovascular Disorders
Research in General Cardiovascular Disorders
Research Targeting Ischemic Heart Diseases
AI in Noninvasive Evaluations
AI During Invasive Procedures
AI in Risk Assessments
Research Targeting Heart Failure
AI in Diagnosing Heart Failure
AI to Predict Prognosis of Heart Failure
Research Targeting Arrhythmias
AI to Diagnose Arrhythmias
AI to Monitor for Arrhythmias
AI to Identify Arrhythmias from Sources Other than ECGs
Discussion
References
59 AIM in Medical Robotics
Introduction
Pre-operative Planning
Pre-/Intra-operative Registration
Execution
Intra-operative Image Analysis
Monitoring and Assessment
Conclusion
Cross-References
References
60 AI in Surgical Robotics
Introduction
Cognitive Surgical Robots
Proprioception
Depth Perception
Navigation
Surgical Tool Tracking
Haptic Feedback and Tissue Interaction Sensing
Advanced Visualization with Augmented Reality
Robot-Assisted Task Execution
Context-Aware Decision Support
Outlook
References
61 Artificial Intelligence in Surgery
Introduction
AI-Powered Techniques
Computer Vision
Technical Aspects of Computer Vision in Surgery
Computer Vision and Supervised Learning
Computer Vision and Unsupervised Learning
Natural Language Processing
Current Applications of AI in Surgery
Preoperative Risk Prediction
Intraoperative Video Analysis
Surgical Workflow Analysis
Regulatory and Legal Considerations
Conclusion
References
62 Artificial Intelligence in Urology
Introduction
Artificial Intelligence in Urology
Urologic Oncology
History
Prostate Cancer
Kidney Cancer
Urothelial Cancer
Endourology
Andrology
Prediction of Male Reproductive Potential
Semen Analyses
Predicting Sperm Retrieval Success Rates
Predicting Surgical Shunt Intervention for Priapism Management
Conclusion
References
63 Artificial Intelligence in Trauma and Orthopedics
Introduction
Diagnostics
Musculoskeletal Image Scheduling and Protocoling
Musculoskeletal Image Acquisition
Musculoskeletal Image Interpretation
Fracture Detection
Knee Pathology Detection and Segmentation
Osteoarthritis Detection and Cartilage Segmentation
Orthopedic Implant Detection
Orthopedic Oncology Detection
Intraoperative and Robotics
Semiautonomous Intraoperative Robotics
Autonomous Intraoperative Robots
Continued Adoption of Robotics
Predictive Analytics
Orthopedic Databases
Predicting Disease Onset and Degree
Postoperative Complications and Rehabilitation
Conclusion
References
64 Harnessing Artificial Intelligence in Maxillofacial Surgery
Introduction and Background
The Maxillofacial Surgeon and AI
Is AI a Friend or Foe?
Simplifying AI for the Surgeon: Suggestions for Seamless Integration of AI and Surgery
Machine Learning and Deep Learning
Artificial Neural Networks
Natural Language Processing
Computer Vision
Surgeon Dilemmas on AI
Role of Surgeons in Enabling AI-Assisted Maxillofacial Surgeries
Challenges in the Path
Suggestions for Overcoming This Challenge
Suggestions for Overcoming This Challenge
Suggestions for Overcoming This Challenge
Literature Speaks: An Overview of Current Applications with Potential for Harnessing AI in Maxillofacial Surgery
Maxillofacial Presurgical Imaging
Current
Potential
Orthognathic Surgery
Current
Potential
Implant Surgery
Current
Potential
Temporomandibular Joint (TMJ) Surgery
Current
Potential
Oncosurgery and Reconstruction
Current
Potential
Trauma Surgery
Current
Potential
Impacted Teeth and Minor Oral Surgery
Current
Potential
Miscellaneous
Watch List for Future Forward Maxillofacial Surgeons
Surgical Data Science
Surgical Scene Analytics
Surgical Control Tower (SCT)
Conclusion
References
65 AIM in Dentistry
Introduction
AI for Assessment in Dentistry
AI for Diagnosis in Dentistry
AI for Treatment Planning
AI for Outcome Prediction in Dentistry
Concluding Remarks
References
66 Artificial Intelligence in Gastroenterology
Introduction
GI Endoscopy
Existing Methods
Hand-Crafted-Feature-Based Approaches
Deep Learning-Based Approaches
Unsupervised and Semi-supervised Approaches
Example Results
Open Issues and Ongoing Research
Limited Data Availability
Generalizability
Metrics and Evaluation
Automatic Report Generation
Explainability
Competitions and Challenges
Clinical Verification and Emerging Commercial Systems
Summary and Conclusions
References
67 AIM in Endoscopy Procedures
Introduction
Applications of Artificial Intelligence to Endoscopy Practice
Detection and Diagnosis During Endoscopic Procedure
Informative Frame Selection
Mosaicking and Surface Reconstruction
Augmented Reality Systems for Intraoperative Assistance and Surgeon Training
Discussion and Perspectives
Conclusion
References
68 AIM in Barrett´s Esophagus
Introduction
Problem Statement
The Case for AI in the Esophagus
AI for Barrett´s Esophagus
BE Cancer Detection Using White Light Endoscopy
BE Cancer Detection Using Narrow-Band Imaging
BE Cancer Detection Using Endomicroscopy
AI for Quality Assessment in the Esophagus
Discussion
References
69 Artificial Intelligence for Colorectal Polyps in Colonoscopy
Introduction
Components for Developing DL
Datasets
Polyp Detection and Localization
Polyp Segmentation
Polyp Classification
Conclusions and Future Trends
Cross-References
References
70 AIM in Otolaryngology and Head and Neck Surgery
Introduction
A Brief Introduction into Machine Learning
Artificial Intelligence in ENT
Head and Neck Cancer
Radiomics
Radiological Staging
Clinical Head and Neck Oncology
Histopathology
Multispectral Imaging
Genetics and Molecular Markers
Thyroid and Endocrine Surgery
Thyroid Cancer
Parathyroid Surgery
Otology
Imaging Modalities and Radiomics
Auditory Brainstem Response Interpretation
Sensorineural Hearing Loss Prediction
Hearing Impairment Technologies
Balance and Vestibular Pathologies
Rhinology
Imaging Diagnosis
Pathological Diagnosis
Chronic Rhinosinusitis Endotyping
Laryngology
Voice and Larynx
Swallow
Limitations, Challenges, and the Future
Conclusion
References
71 AIM in Obstetrics and Gynecology
Introduction
AI in IVF
Ethical Challenges
References
72 AIM in Medical Disorders in Pregnancy
Introduction
Artificial Intelligence in Reproductive Medicine
Opportunities and Limitations of AI in Reproductive Medicine
AI for Assessment, Diagnosis, or Treatment of Infertility
AI for Embryo Annotation, Evaluation, and Selection
AI for Prediction of Embryo Chromosome Status (Ploidy)
AI in Maternal Healthcare Miscarriage Prediction
Conclusion
References
73 AIM and Gender Aspects in Reproductive Medicine
Introduction
The History of Reproductive Medicine
The Relationship Between AI and Reproductive Medicine
Gender and Reproductive Medicine: Objectives, Issues, and Challenges
Reproductive Medicine and Gender: A Century-First Century Approach
The Male Experience and Reproductive Medicine
Transgender, Gender Fluid, and Nonbinary Perspectives and Reproductive Medicine
Concluding Thoughts
References
74 Artificial Intelligence in Pediatrics
Introduction
Challenges in Pediatric AI
Recent Developments in Pediatric AI
Cardiology
Respiratory
Genetics
Endocrinology
Neonatology
Neonatal Sepsis
Jaundice
Ophthalmology
Primary Care
Radiology
Pediatric Intensive Care
Gastroenterology
Future Potential for AI in Pediatrics
References
75 AIM in Neonatal and Pediatric Intensive Care
Introduction
Sepsis Definition
Continuous Vital Sign Assessment to Predict Life-Threatening Events
Neonatal Sepsis and the NICU
Pediatric Sepsis and Early Detection
Challenges and Future Perspectives of Automated Vital Signs Pattern Analysis
Conclusions
References
76 Aging and Alzheimer´s Disease
Introduction
AI in Healthcare
Historical Overview
From the 1950s to the 2000s
After the 2000s
AI for Drug Discovery
Virtual Screening
Bioactivity Scoring
ADME/T Properties Prediction
De Novo Drug Design
AI in Biology
Genetics
Proteomics
AI in Medicine
Diagnosis
Prognosis
Aging Biomarker Development
Application of Machine Learning in Clinical Work for Alzheimer´s Disease
Etiology
Diagnosis
Therapy
Prognosis
Future Perspectives and Concluding Remarks
References
77 Aim in Genomics
Network Medicine: A New Paradigm for the Study of Diseases
An Introduction to Network Medicine
Machine Learning Challenges in Network Medicine
A Network-Based Analysis of Disease Modules Using a Taxonomic Perspective
Construction of the Interactome Taxonomy (I-T)
Taxonomy Alignment and Labeling
Taxonomy Alignment
Comparing Alternative Induced Taxonomies
Interactome Hierarchy (I-T) Labeling
Experimental Set-up
Discussion
Finding Disease Categories with a Corresponding Dense Neighborhood in the Interactome
Finding Unexpected Structural Relations Between Disease Categories
Detection of Nomenclature Errors in Disease-Gene Associations
Conclusions
References
78 AIM in Genomic Basis of Medicine: Applications
Introduction
Classification of Genomic Variation
Variants in the Coding Region
Variants in the Non-coding Region
Interpretation of Variants Using NLP
Diagnosis (Phenotyping)
Proposal for Optimal Drug Treatment
Summary and Future Implications
References
79 Stem Cell Progression for Transplantation
Introduction
Implementation of AI in Stem Cell Progression
Machine Learning
Computational Methods
Deep Learning
Conclusions
References
80 Artificial Intelligence in Blood Transcriptomics
Introduction
Blood Transcriptomics in Clinics: Methods, Features, Pitfalls
Background on Artificial Intelligence in Biology
Research Development Towards Clinical Applications
Ethical Considerations, Data Security and Federated Learning
Outlook
Cross-References
References
81 AIM in Health Blogs
Introduction
Related Studies
Social Analytics for Healthcare
Adverse Drug Reaction (ADR)
Methodology
Data Gathering
Data Filtering
Topic Modeling
Dimensionality Reduction
Agglomerative Hierarchical Clustering
Summarization
Evaluation and Experiments
Dataset and Diseases
Compared Systems
Evaluation Strategy
Topic Analysis and Subclustering
Data Analytics
Conclusion
References
82 AIM and Transdermal Optical Imaging
Introduction
Characterizing the Cardiovascular System: Benefits, Obstacles, and Breakthroughs
Blood Pressure Measurement
Hypertension Screening, Diagnosis, and Management
General Health Assessment
General Health Monitoring
Heart Rate, Heart Rhythm, and Heart Rate Variability
General Assessment and Monitoring
Stress Assessment
The Need for Technological Breakthrough
Transdermal Optical Imaging
Overcoming Measurement Obstacles with Transdermal Optical Imaging Technology
Scientific Foundations of Transdermal Optical Imaging
Biomechanics and Video Capture
Extracting Plethysmographic Signal from Video Using Artificial Intelligence
Cardiovascular Parameters and Their Relation to Blood Flow Features
Pulse Rate
Pulse Rate Variability
Blood Pressure
Predicting Blood Pressure Through the Efficient Combination of Feature Information Using Artificial Intelligence
Present Advances Using Transdermal Optical Imaging
Accurate Blood Pressure Measurement
Accurate Heart Rate and Heart Rate Variability Measurement
Trends in Medicine and the Potential Impact of TOI
Growing Challenges for Healthcare Delivery
Improving Healthcare Quality and Efficiency with Patient-Centered Health Monitoring
Improving Healthcare Accessibility with Telemedicine
TOI-Based Tools Could Transform Personalized Self-Monitoring and Telemedicine
Future Uses and Challenges for TOI
References
83 AI in Longevity Medicine
Introduction
The Advent of Deep Aging Clocks
Federated Learning for Biomarker Discovery and Development
Longevity Physicians: Emerging Specialists and the Need of a Tailored Education
Healthy Versus Wealthy Longevity: Longevity Medicine and Public Health
AI Applications in Medicine: The Fundament of Precision Medicine
Conclusion and Future Perspectives
References
84 AIM in Nanomedicine
Introduction
Review of Literature
Results and Discussion
Machine Learning for Nanodrug Discovery and Nanoformulations
Drug Delivery Systems and Formulation
Machine Learning in Specific Areas of Nanomedicine
Mathematical Machine Learning Modeling for Cancer Nanomedicine and Theranostics
Machine Learning in Precision Nanotheranostics for Cancer
Machine Learning for Nanotoxicology
Machine Learning and Quantum Enabled Technologies
Machine Learning and Regenerative Nanobiology
Next Generation Machine Learning for Nanorobotic Surgery
Ethics and Regulation
References
85 AIM in Wearable and Implantable Computing
Introduction
Context Awareness
Definition and Theory
Context Types and Recognition Path
Selected Recognition Problems
Context Pattern Spotting and Interpretation
Interpretation from Context Hierarchy
Flu Detection
Situation Interpretation in Implants
Design and Construction
Topology Design and Optimization
Wearable Personalization
Prefabrication and Process Planning
AI-Assisted Fabrication of Wearables
Validation
AI System safety assessment
Implantable System Testing
Self-Checking AI Systems
Usability
References
86 Machine Learning and Electronic Noses for Medical Diagnostics
A Brief Introduction to the Electronic Nose Technique
An Overview of Electronic Nose Application in Medical Diagnostics
Machine Learning in Enhancing the Diagnostic Capability of Electronic Noses
Denoising Algorithms
Sensor Drift Compensation
Conclusions
References
87 Artificial Intelligence in Telemedicine
Introduction
The Basics of Telemedicine
The Integration of Artificial Intelligence in Telemedicine
Telemedicine and Artificial Intelligence
Teleophthalmology and AI
Telestroke and AI
Teledermatology and AI
Telemedicine, Artificial Intelligence, and Education
References
88 AIM and mHealth, Smartphones and Apps
Introduction
History
AI and mHealth in Various Medical Specialties
AI and mHealth for Evidence-Based Medicine
AI and mHealth in the Field of Genomics
AI and mHealth in the Field of Cardiovascular Medicine
AI and mHealth in the Field of Respiratory Medicine
AI for mHealth for Neuroscience and Neuropsychiatric Disorders
AI in mHealth for Rheumatology
AI in mHealth for Gastroenterology
AI in mHealth for Urology
AI in mHealth for Endocrinology
AI in mHealth for Dermatology
AI in mHealth for Obstetrics and Gynecology and Pediatrics
AI in mHealth for Consensus Evaluation
AI and mHealth in Infectious Diseases
Summary and the Future of mHealth
References
89 AIM in Alternative Medicine
Introduction
Methodological Approaches
Data Sources
Knowledge Engineering Approaches
Expert System
Ontology and Knowledge Graph
Machine Learning and Data Mining Approaches
Applications and Related Work
Clinical Diagnosis
Clinical Therapies
Pharmacological Applications
Biomechanisms of Syndrome
Shortcomings and Future Directions
Conclusion
References
90 AIM in Oncology
Introduction
AI Tools
Cancer Detection
Cancer Treatment
Conclusion
References
91 Artificial Intelligence in Radiotherapy and Patient Care
Introduction
Basic Concept of AI and Machine Learning
Radiotherapy Chain
Applications of AI and Machine Learning in Radiotherapy
Radiation Treatment Planning
Treatment Plan Evaluation
Treatment Plan QA
Dose Distribution Index Prediction
Radiation Dose Delivery Using Multileaf Collimator
Chatbot
Background and Basic Concept of a Chatbot
Chatbot for Radiotherapy Education and Patient Care
Conclusion and Future Prospective
References
92 Deep Learning in Mammography Breast Cancer Detection
Introduction
Datasets
MIAS
DDSM
INbreast
OPTIMAM Mammography Image Database (OMI-DB)
BCDR
Deep Learning Methods
Performance Metrics
Discussion and Future Challenges
Conclusion
References
93 AIM for Breast Thermography
Introduction
Conventional Breast Imaging Techniques
X-Ray Mammography
Ultrasound
Magnetic Resonance Imaging
Challenges with Conventional Breast Imaging Modalities
Infrared Thermography for Breast Imaging
Breast Thermal Imaging Protocol
Challenges with Manual Interpretation of Breast Thermography
Artificial Intelligence for Breast Thermography
AI for View Labeling
AI for Breast Segmentation
AI for Malignancy Classification
AI for Risk Estimation
AI for Biomarkers Prediction
Discussion
Conclusion
References
94 AIM and Cervical Cancer
Introduction
Cervical Cancer and the Role of Pap Smear Test in Early Detection
Pros and Cons of the Automated System of Diagnosis Concerning the Manual Diagnosis
Challenges and Advances in Automation of Pap Smear Images
Line of Action or AI-Driven Approaches for the Automation Task Using Pap-Smear Images
Pap Smear Image Database Generation
Ground Truth Labeling
Automated Cell or Nuclei Segmentation from Whole Slide Image
Feature Extraction and Feature Selection Methods
Automated Binary or Multi-Class Classification of Pap Smear Images
Summary
References
95 Artificial Intelligence in Infectious Diseases
Introduction
Applications of Artificial Intelligence in Infectious Diseases
Artificial Intelligence for the Identification of Microorganisms
Microorganisms Detection, Identification, and Quantification
Evaluation of Antimicrobial Susceptibility
Diagnosis, Disease Classification, and Clinical Outcomes
Artificial Intelligence for the Clinical Diagnosis and Management of Infectious Diseases
Early Detection and Management of Sepsis
Diagnosis of Infection
Prediction Tools
Antimicrobial Selection
Artificial Intelligence in Surveillance and Infection Prevention
Challenges and Limitations in the Development and Application of Artificial Intelligence in Infectious Disease Management
Development, Implementation, and Adoption
Conclusion and Recommendations for Artificial Intelligence in Infectious Diseases
References
96 Artificial Intelligence in Epidemiology
Introduction
Present Advances
Collecting Data for AI Development and Developping AI for Data Collection
AI to Collect, Classify, and Structure Data
AI to Reconstruct or Virtualize Experimental Designs
Disease and Health Outcomes Surveillance
Pharmacovigilance
Sentiment Analysis and the Use of Social Media as Alternative Data Source and Data Processing Method in Epidemiology
Recruitment of E-cohorts and Online Syndromic Surveillance
Social Media Content and Sentiment Analysis
From Data to Decision: Data- and Model-Driven Knowledge and Decision in Epidemiology
Potential Trends and Future Challenges
The Challenges Related to Data
The Challenges Related to Epidemiology as an Explanatory Science: Statistical Inference and Causality
The Challenges Related to the Use Made of Epidemiology by Decision-Makers and the Involvement of Private Actors
Conclusion
References
97 Artificial Intelligence and Malaria
Introduction
Malaria Diagnosis: Parasite Detection and Species Identification
Vector Ecology: Species, Biology, and Behaviors
Deployment of Artificial Intelligence in Malaria
Applications of Artificial Intelligence in Malaria Diagnosis and Malaria Vector Characterization
Image-Based Automatic Classification for Malaria Diagnosis
Plasmodium falciparum Detection
Determination of Plasmodium Life Stages
Mobile Applications and End-to-End Systems for Parasite Identification and Density Determination
Image-Based Automatic Classification of Mosquito Vectors
Mosquito Species Identification
Mosquito Behavioral Patterns
Characterization of Anopheles Biological Features Using Proteomic Tools
Conclusion
Cross-References
References
98 Artificial Intelligence in Infection Biology
Introduction
Aim in the Infection Biology
Computer Vision in Infection Biology on Nano- and Microscale
Computer Vision in Infection Biology on Mesoscale and Aspects of Temporal Dimensions
Computer Vision in Infection Biology on Macroscale and Digital Biomarkers
Artificial Intelligence in Molecular Infection Biology
Conclusions
References
99 Artificial Intelligence in Medicine: Modeling the Dynamics of Infectious Diseases
Epidemiology
History
Using Artificial Intelligence
The Recurrent Neural Network Approach
The Multi-agent Approach
SARS-CoV-2
Computing Requirements
Parameter Uncertainty
Conclusion
References
100 AI and Immunoinformatics
Introduction
AI and Vaccine Discovery
AI and Vaccine Discovery for SARS-CoV-2
AI for MHC Binding Peptide Prediction and Organ Transplant
The Challenge of Variation
Drug Discovery and Vaccine Discovery Differences for Learning
Discussion
References
101 Artificial Intelligence in Clinical Immunology
Introduction
Healthcare Data as Big Data
Structuring Healthcare Data
What Is AI and How Does It Work?
AI and the Learning Health System
Applications of AI in Clinical Immunology
Disease Diagnosis: Primary Immunodeficiency
Clinical Decision Support
Candidate Selection for Clinical Trials and Patient Identification Efforts
COVID19 Applications and Impact
Biomarker Identification in Clinical Immunology
Microbiome Analysis
Cytometric Analysis
AI Governance for CI
Ethical Implications in AI
Conclusions
References
102 AIM in Allergy
Introduction
General Principles of Machine Learning
The Allergic March: Can We Predict the Direction?
The Needle in the Hay: ML-Approaches for Allergy Biomarker Discovery
Finding a Pattern: Disease Subtyping for Clinical Management
Bridging the Gap: Insights into Disease Biology by Integrative Analysis
Challenges and Promising Research Directions in ML Applications
References
103 AIM in Haematology
Introduction
Discussion
History
Artificial Intelligence in Diagnostic Haemopathology
Use of Neural Networks for Peripheral Blood Film Analysis
Classification of White Blood Cells Using Transfer Learning
Screening Blood Disorders like Thalassemia
Screening for Polycythemia Rubra Vera
Screening for Pyruvate Kinase Deficiency
Digital Morphological Analysis
Artificial Intelligence in Haem-oncology, Cancer Stem Cells, and Cancer Immunotherapy
AI for Haem-oncology Screening
AI for Immune Checkpoint Blockade Discovery
Leukaemia and Lymphoproliferative Disorders
AI for Acute Myeloid Leukaemia
AI for Acute Promyelocytic Leukaemia
AI for Lymphoma
AI in Acute Lymphoblastic Leukaemia
AI in Multiple Myeloma and Cancer Stem Cells
AI in Neoplastic Bone Marrow Disease
AI for Lymphoproliferative Disorders
AI for Cancer Immunotherapy
AI for Haematological Gene Profiling and Molecular Sequencing
AI for Cancer Disease Progression Monitoring
AI for Haemo-virology, Haemo-parasitology, and Immune System Studies
AI in COVID-19 Haemo-virology
AI in HIV and AIDS Haemo-virology
AI in Haemo-parasitology
AI for Haemorrhage Prediction and Transfusion
AI for Transfusion and Blood Quality Haemo-diagnostics
AI for Postpartum Haemorrhage Prediction
AI for Bone Marrow Haematopoietic Stem Cell Transplantation
AI in Autoimmunity
Point-of-Care Diagnostics, Precision Diagnostics, and Sports Haematology
AI in Point-of-Care and Precision Haematological Diagnostics
AI in Sports Haematology
Future Considerations for AI in Haematology
References
104 Artificial Intelligence in Medicine in Anemia
Introduction
Artificial Intelligence Tools for Anemia Management
Pathophysiology and Treatment of Anemia in Chronic Kidney Disease
Application of Artificial Intelligence to Diagnosis and Management of Anemia
Expert Systems
Moving Past Just Imitation
Artificial Neural Networks
Reinforcement Learning
Fuzzy Systems
Conclusion
References
105 AIM in Anesthesiology
Introduction
The Use of AI to Monitor Depth of Anesthesia
Controlling Anesthesia Delivery with AI
Perioperative Hemodynamic Optimization Assisted by AI
Automation for Fluid Therapy
Automation for Vasopressor Titration
Automation for Inotrope Infusion
Automation for Vasodilator Infusion
Machine Learning for Predicting Hypotension
Event and Risk Prediction with AI
Other Applications of AI in Anesthesiology
Technology Readiness Level of Published Applications
Implications of AI for the Anesthesiologist
Conclusion
References
106 Artificial Intelligence in Critical Care
Introduction
Interpretation, Explanation, Pipelines, and Guidelines
Where at the ICU Should We Apply AI and ML?
What Sort of AI and ML Can We Apply at the ICU?
A Further Few Things About the Use of AI and ML in Medicine that Merit Discussion
Conclusions
References
107 Artificial Intelligence in Medicine (AIM) for Cardiac Arrest
Introduction
Cause of Cardiac Arrest: Ventricular Arrhythmias
Baseline Cardiovascular Diseases Leading to Ventricular Arrhythmia
From Ventricular Arrhythmias to Cardiac Arrest
The Current Clinical Treatment to Prevent Cardiac Arrest
The Possibility and Advantage to Introduce AI Technology
Overview of Researches to Prevent Cardiac Arrest Utilizing AI
Early Recognition/Detection of High-Risk Patients
Active Intervention and Follow-up
Continuous Monitoring and Subsequent Interventions
Discussion
References
108 Artificial Intelligence in Clinical Toxicology
Introduction: Clinical Toxicology
The Importance of Toxicovigilance
Predicting Clinical Efficacy and Drug Toxicity
Artificial Intelligence
Machine Learning Algorithms
Supervised Learning
Unsupervised Learning
Artificial Neural Networks (ANNs)
Deep Learning Algorithms
Advances in Computational Toxicology
Big Data for Toxicology Interpretations
Physiological-Based Pharmacokinetic (PBPK) Modeling
Conclusion
References
109 Artificial Intelligence in Acute Ischemic Stroke
Introduction
AI Applications to Acute Stroke Medicine
Diagnosis
Lesion Segmentation
Clot Detection
Prediction
Imaging Outcomes
Clinical Outcomes
Integration
Challenges for AI in Stroke Medicine
The Black Box Problem
Evaluation of AI Models
Data Registries
Conclusion
References
110 Artificial Intelligence and Deep Learning in Ophthalmology
Introduction
Essential Concepts and Components in an AI System
Application of AI and DL Algorithms in Ophthalmology Using Different Devices
Retinal Fundus Photographs
Diabetic Retinopathy
Glaucoma
Age-Related Macular Degeneration
Retinopathy of Prematurity
Papilledema and Optic Disc Abnormalities
Systemic Diseases
Optical Coherence Tomography
DL Algorithms for Retinal Diseases Using Macula-Centered OCT
DL Algorithms for Glaucoma Using Optic Disc-Centered OCT
DL Algorithms for Glaucoma Using Anterior-Segment OCT
Visual Fields
Infantile Facial Video Recording
Electronic Health Records
Image Quality Assessment
Future Research and Challenges
Novel Technical Approaches
Research Ethics and Artificial Images
Data Ownership and Sharing
Patients and Physicians Acceptance
Education
Guidelines
Conclusion
References
111 Artificial Intelligence in Ophthalmology
Introduction
Clinical Application of AI in Ophthalmology
Diabetic Retinopathy (DR)
Glaucoma
Age-Related Macular Degeneration (AMD)
Conclusions
References
112 Aim in Depression and Anxiety
Introduction
AI and Machine Learning for Precision Medicine in Depression and Anxiety Disorders
Aims for Using AI in Depression and Anxiety
Risk
Diagnosis
Treatment Outcome
Suicidality
Relapse
Outlook
Cross-References
References
113 Artificial Intelligence for Autism Spectrum Disorders
Introduction
AI Applications for ASD: Objectives, Data, and Challenges
Objective-Based Categorization
Automatic/Early Diagnosis
Severity Recognition
Subtypes Definition
Longitudinal Studies
Explorative Analysis
Drug Discovery
Teaching and Interaction
Field-Based Categorization
Brain Structural Imaging
Brain Functional Imaging
Genetics
Video and Sensor Analysis
Miscellaneous
Challenges of the AI-Based Research on ASD
Conclusions
Cross-References
References
114 Artificial Intelligence in Schizophrenia
Introduction
Artificial Intelligence Applied to the Research, Diagnosis, and Treatment of Schizophrenia and Related Disorders: Pre-2000
Artificial Intelligence Applied to the Research, Diagnosis, and Treatment of Schizophrenia and Related Disorders: 2000-2012
Artificial Intelligence Applied to the Research, Diagnosis, and Treatment of Schizophrenia and Related Disorders: 2012-2018
Artificial Intelligence Applied to the Research, Diagnosis, and Treatment of Schizophrenia and Related Disorders: 2019-Present
Current and Future Clinical Use of AI Techniques in the Diagnosis and Treatment of Schizophrenia and Related Disorders
Cross-References
References
115 The Rise of the Mental Health Chatbot
Introduction
Mental Health Support using Chatbots
Chapter Summary
Mental Health Chatbots
The Value of Mental Health Chatbots
How Chatbots Work
Economic Impact
How a Chatbot Could Change the Economics of the Employee Benefits Industry
Factors Causing Low Utilization
Utilization Strategies of Change
Factors Causing Low PEPM
PEPM Strategies of Change
Real-World Applications
Case Study: 24/7 Access for 430,000 People Served by a Public Health Department
Case Study: AI Mental Health Support for Young Mothers in Africa
Case Study: Supporting Caregivers and Their Patients
Case Study: Addressing Comorbidities in Childhood Obesity, Prediabetes, and Mental Health Struggles
References
116 AIM in Alcohol and Drug Dependence
Introduction
Challenges in Diagnosis and Treatment
Role of Artificial Intelligence
Data Sets for Machine Learning
Machine Learning for Drug Dependence
Challenges and Outlook
Conclusion
Cross-References
References
117 Artificial Intelligence in Medicine and PTSD
PTSD: A Complex Clinical Disorder
PTSD: A Psychological Disorder that May Appear After Exposure to a Traumatic Event
PTSD: A Disorder that Is Currently Difficult to Explain and Predict
The Evolution of PTSD Over Time May Be Complex and Chronic, and Associated with Physical and Mental Disorders of a Different N...
Improving Screening for and Diagnosis of PTSD
Present Advances: AI and PTSD
AI and PTSD Prediction. AI for Clinical Practice and Practitioners
Early Prediction of PTSD
Prediction of Response to Treatment
AI, Characterization and Diagnosis of PTSD. AI for Basic Research
The Use of Genomic and Neuroimaging Data by AI
Genomic Data
Neuro-Imaging Data
AI for the Diagnosis and Differentiation of PTSD from Other Mental Illnesses
The Use of AI to Characterize Subtypes or Subsyndromic Forms of PTSD
The Contribution of AI in Linking Basic Research to Clinical Applications. Neuroimaging Data
Potential Trends and Future Challenges
References
118 AIM in Eating Disorders
Introduction
AI for ED
Monitoring of Dietary Behavior
Methods in Automated Dietary Monitoring (ADM)
Wearable-Based ADM
Smartphone-Based ADM
Ambient Technology-Based ADM
Digital Biomarkers for AIM in EDs
Intake Timing
Food Type
Image-Based Food Type Recognition
Audio-Based Food Type Recognition
Evaluation Metrics for Food Type Recognition
Food Amount
Intake-Accompanying Phenomena Related to EDs
Triggers and Stressors
Example: AI in Anorexia Nervosa (AN)
Prospects for AN
References
119 AIM in Neurology
Introduction
Childhood Medulloblastoma
Methods and Materials for AI Application
Significance of an Accurate Detection in Prognosis
Challenges of a Clinical Observation
Advantages of AI-Driven Solution
Challenges of an AI-Driven Solution
Scope for Industry Transformation
Summary
References
120 AIM in Neurodegenerative Diseases: Parkinson and Alzheimer
Introduction
Artificial Intelligence for Dementia
AI for Dementia Diagnosis Using Big Data
Conditional Restricted Boltzmann Machines in Alzheimer´s Disease
Computer Vision for Dementia Patient Video Monitoring and Analysis
AI and Assistive Robotic Technologies for Dementia
Cognitive and Behavioral Biomarker, Facial Motion Assessment Using AI
Dementia-Related Electroencephalographic Analysis and Robotic-Assisted AI
Vascular Dementia and AI
Convolutional Neural Nets and Model Explainability in Dementia AI Studies
Artificial Intelligence in Parkinson´s Disease
AI in Lewy Body Dementia
Motor and Gait Impairment Detection Using AI
AI for Electroencephalographic Diagnosis and Prognostication in Parkinson´s Disease
AI for Parkinson´s Disease Medical Management Drug Repurposing
AI for Parkinson´s Disease Surgical Management
Ethical and Social Implications of AI for Parkinson´s and Dementia
Future In Vivo Detection and Management of Dementia and Parkinson´s Using Quantum AI Systems
References
121 AIM in Amyotrophic Lateral Sclerosis
Introduction
Review
Clinical Trial Analysis of ALS Disease
PRO-ACT Dataset
Study of ALS with Machine Learning Approach
Experimental Results
Conclusion
References
122 AIM in Ménière´s Disease
Introduction
Overview of Ménière´s Disease
The History of Inner Ear MRI to Visualize Endolymphatic Space and Hydrops
Current Diagnostic Method and Dilemma in Ménière´s Disease
Sequence and Analysis of Inner Ear MRI for the Diagnosis of Ménière´s Disease
Development of Artificial Intelligence in the Medical Field: Focusing on Medical Image Analysis
The Use of Artificial Intelligence in Ménière´s Disease
The Future of Artificial Intelligence in Ménière´s Disease
References
123 AIM and Brain Tumors
Introduction
Magnetic Resonance Imaging of Brain Tumors
Automated Analysis of MRI in Brain Tumors: Challenges
Structure of the Chapter
Brain Tumor Detection and Segmentation
Automatic Detection and Segmentation of Brain Tumors
Dealing with Limited Ground-Truth Data Sets
Assessing Automatic Detection and Segmentation
Analysis of Segmented Brain Tumors
Quantification of Tumor Characteristics
Classification of Brain Tumors Using AI
Conclusion
References
124 Artificial Intelligence in Stroke
Introduction
Artificial Intelligence
Early Detection of Stroke Symptoms
Acute Stroke Therapy
Role of AI in the Subacute Phase and Follow-Ups of the Ischemic Stroke Patients
Use of AI in the Management of Transient Ischemic Attack
Role of AI in the Management of Intracerebral Hemorrhage
Future Directions
References
125 AIM in Clinical Neurophysiology and Electroencephalography (EEG)
Introduction
Epilepsy
Artificial Intelligence, Machine Learning, and Deep Learning
Machine Learning and Deep Learning Approaches in Epilepsy
The Role of History-Taking and Where Deep Learning Can Make In-Roads
Deep Learning Approaches for Investigating Epilepsy
Machine Learning for EEG Analysis
Deep Learning in Epilepsy Treatment
Machine Learning for Epilepsy Surgery
Machine Learning for Electrophysiological Migraine Detection
Deep Learning for Electromyography
Practical Machine Learning for EEG Analysis
Concluding Remarks
Highlighting the Tension Between Progress and ``Model Explainability´´
References
126 Artificial Intelligence in Forensic Medicine
Introduction
Evidence and Individualization in Forensic Medicine
Data, Information, and Evidence in Medicine and in Law: AI at Every Step in the Process
Personalized Medicine, Individualized Sentences: A Shift from the Group to the Individual in Society
Forensic Reasoning
Artificial Intelligence and Assisted Decision-Making in Forensic Medicine
AI and Clinical Forensic Medicine
AI and Forensic Pathology
AI and Medical Expertise in the Legal Context
AI on the Borders of Forensic Medicine
Artificial Intelligence for Forensic Medicine Research
Potential Trends and Future Challenges
Cross-References
References
127 AI in Forensic Medicine for the Practicing Doctor
Introduction
AI Applications for the Forensic Medicine Practice
Part A: Description of AI Applications for the Forensic Medical Doctor
Thanatology
Postmortem Identification
Postmortem Interval (PMI) Estimation
Determination of the Causes of Death
Clinical Forensic Medicine
Part B: Applicability and Usefulness of AI in Current and Future Practices in Forensic Medicine
Conclusion
References
128 Artificial Intelligence for Physiotherapy and Rehabilitation
Introduction
AI in Physiotherapy and Physical Rehabilitation of Patients
AI in Exergames and Serious Games for Early- Stage Physical Rehabilitation
AI in Physiotherapy Education and Use of Simulation for Educating Physiotherapists
AI and Physiotherapy Education
AI for Robotic Assisted Physiotherapy
AI for Physio-assisted Activity of Daily Living Monitoring
AI and Virtual Reality for Physiotherapy and Rehabilitation
AI for Physiotherapy-assisted Sensory and Balance Training
AI for Assisted Wheelchair Users and Assisted Mobility Support
AI for Inattention and Hemi-neglect Training
AI for Respiratory Physiotherapy Management
AI for Community Physiotherapy and Care
AI for Cognitive Impaired Patients Needing Physiotherapy and Rehabilitation
AI for Functional and Feedback Systems in Physiotherapy
AI in Smart Watches and Wearables for Physiotherapy
Future of AI in Physiotherapy
References
129 AIM in Rehabilitation
Introduction
Definition of Rehabilitation
Benefits of Rehabilitation
Phases of Rehabilitation
Areas of Rehabilitation Medicine
Physical Therapy
Speech Therapy
Neuropsychology
Occupational Rehabilitation
General Applications of AI in Rehabilitation
Robotics in Rehabilitation
Learning from Demonstration
Gait Rehabilitation
AI in Assessment and Decision Support Systems
AI in Rehabilitation Prognosis
AI Wearable Monitoring Devices
Fall Detection
Monitoring Purposes
AI in Virtual Reality and Serious Games
Summary
References
130 AIM in Sports Medicine
Introduction
Advances
Potential Trends
Future Challenges
References
Index
