AI in Clinical Medicine: A Practical Guide for Healthcare Professionals

Cover
Title Page
Copyright Page
Dedication
Contents
List of Contributors
Foreword
Preface
Acknowledgements
Relevant AI Terms
About the Companion Website
SECTION I Overview of Medical AI: The What, the Why, and the How
Chapter 1 An Introduction to AI for Non-Experts
1.1 Introduction
1.2 Machine Learning
1.3 Strategies to Train Algorithms
1.4 Underfitting and Overfitting
1.5 Data Preparation and Its Importance
1.6 Artificial Neural Network
1.7 Training a Neural Network: All You Need to Know
1.8 Advanced Techniques
Chapter 2 General Framework for Using AI in Clinical Practice
2.1 Introduction
2.2 AI in Clinical Use Framework
2.3 Patient Physician Trust
2.4 Regulation
2.5 Evidence and Bias
2.6 Automation
2.7 Ethics and Liability
2.8 Reimbursement
2.9 Equity
References
Chapter 3 AI and Medical Education
3.1 Introduction
3.2 Competency in AI: Preparing for the Era of AI in Medicine
3.3 AI Tools to Enhance Medical Education Itself
3.4 Perspectives and Best Practices in Implementing Changes to the Medical Curriculum
3.5 Future Directions
Disclosure Statement
References
SECTION II AI Foundations
Chapter 4 History of AI in Clinical Medicine
4.1 Introduction
4.2 The Beginnings of AI in Medicine
4.3 The Development of Artificial Neural Networks
4.4 The Era of Support Vector Machines and Feature Descriptors
4.5 Dominance of Deep Learning
4.6 Interpretability of Deep Models
4.7 Transformers: Better Performance and Inherent Interpretability through Self-Attention
4.8 The (Near) Future
Funding
References
Chapter 5 History, Core Concepts, and Role of AI in Clinical Medicine
5.1 Introduction
5.2 Core Concepts of AI
5.3 Roles of AI in Relation to Physicians
5.4 Concluding Remarks
References
Chapter 6 Building Blocks of AI
6.1 Introduction and AI Definitions
6.2 Challenges and Failures in Healthcare AI from a ‘Building Blocks’ Perspective
6.3 Conclusion
References
Chapter 7 Expert Systems for Interpretable Decisions in the Clinical Domain
7.1 Introduction
7.2 Deep Learning in Clinical Medicine
7.3 Interpretability versus Explainability
7.4 Expert Systems in the Clinical Domain
7.5 Roadmap to Knowledge-Driven Deep Learning
7.6 Potential Applications of Mixed Intelligence Systems
7.7 Conclusion
References
Chapter 8 The Role of Natural Language Processing in Intelligence-Based Medicine
8.1 Introduction
8.2 Introducing a Few Common and Important Natural Language Processing Techniques in Healthcare
8.3 Summary of State-of-the-Art Platforms for Natural Language Processing in Healthcare
8.4 Use Case Project for Natural Language Processing – Approaches and Challenges
8.5 Conclusions
Conflicts of Interest
References
SECTION III AI Applied to Clinical Medicine
Frontline Care Specialties
Chapter 9 AI in Primary Care, Preventative Medicine, and Triage
9.1 Introduction
9.2 Primary Care
9.3 Preventative Medicine
9.4 Triage
9.5 Challenges
Conflicts of Interest
References
Chapter 10 Do It Yourself: Wearable Sensors and AI for Self-Assessment of Mental Health
10.1 Introduction
10.2 Psychological Health Assessment Tools
10.3 Modalities for Mental Health Assessment
10.4 Machine Learning for Mental Health Diagnosis – Wearable Sensors and AI
10.5 Discussion and Future Roadmap
References
Chapter 11 AI in Dentistry
11.1 Introduction
11.2 Current Trends: What Is the Focus of Research?
11.3 Use Cases and Application Opportunities
11.4 AI and Dentistry Regulations in the USA
11.5 Challenges
11.6 Roadmap to the Future
Conflicts of Interest
References
Chapter 12 AI in Emergency Medicine
12.1 Introduction
12.2 Prehospital
12.3 Cardiac Arrest
12.4 Triage
12.5 Monitoring
12.6 Electrocardiogram
12.7 Imaging
12.8 Patient Assessment and Outcome Prediction
12.9 Departmental Management
12.10 Public Health: Opioid Overdose and Disease Outbreaks
12.11 Success in Implementing AI in the Emergency Department
12.12 Challenges
12.13 Future Possibilities
12.14 Chapter Limitations
References
Medical Specialties
Chapter 13 AI in Respirology and Bronchoscopy
13.1 Introduction
13.2 Brief Overview of AI and Machine Learning
13.3 Innovative Research Using AI in Bronchoscopy
13.4 Current Clinical Applications of AI in Bronchoscopy
13.5 Final Word
References
Chapter 14 AI in Cardiology and Cardiac Surgery
14.1 Introduction
14.2 Cardiology Imaging and Electrophysiology
14.3 Solutions for Specific Tasks
14.4 Cardiac Surgery
14.5 Remaining Challenges
14.6 Conclusion
References
Chapter 15 AI in the Intensive Care Unit
15.1 Introduction
15.2 AI Methodology in the ICU
15.3 Clinical Scoring Systems
15.4 Improving Sepsis Recognition with AI
15.5 Reinforcement Learning in ICU
15.6 Deep Learning in ICU
15.7 Limitations and Future Directions
Conflicts of Interest
References
Chapter 16 AI in Dermatology
16.1 Introduction
16.2 History of AI in Dermatology
16.3 Potential Clinical Applications
16.4 The Path to Clinical Implementation
References
Chapter 17 Artificial Intelligence in Gastroenterology
17.1 Introduction
17.2 Colonoscopy
17.3 Esophagogastroduodenoscopy
17.4 Video Capsule Endoscopy
17.5 Endoscopic Ultrasound
17.6 Clinical Prediction Models
17.7 Regulation
17.8 Future Directions
Conflicts of Interest
References
Chapter 18 AI in Haematology
18.1 Introduction
18.2 Current Thinking, Testing, and Best Practice in Haematology
18.3 The Implementation of AI in Haematology
18.4 Current Applications of AI in Haematology
18.5 Considerations and Challenges for AI in Haematology
18.6 The Future of AI in Haematology
References
Chapter 19 AI and Infectious Diseases
19.1 Introduction
19.2 Disease Outbreaks and Surveillance
19.3 Disease Diagnosis
19.4 Prediction and Control of Antimicrobial Resistance
19.5 Disease Treatment
19.6 Conclusion
References
Chapter 20 AI in Precision Medicine: The Way Forward
20.1 Precision Medicine: The Future of Healthcare
20.2 The Way Forward
20.3 Benefits of Precision Medicine in the Clinical Care Pathway
20.4 Investing in Precision Medicine–Based Drugs Is the Future of Drug Development
20.5 Healthcare Stakeholders and Their Role in the Success of Precision Medicine
20.6 AI, Big Data, Wearables, and Real-World Data
20.7 Future Perspective
References
Chapter 21 AI in Paediatrics
21.1 Introduction
21.2 The Use of AI in Paediatrics
21.3 Challenges for AI in Paediatric Healthcare
21.4 Future Directions
References
Chapter 22 AI Applications in Rheumatology
22.1 Introduction
22.2 Inflammation
22.3 Damage
22.4 Disease Activity
22.5 Systemic Sclerosis
22.6 Limitations and Future Perspectives
References
Surgical Specialties
Chapter 23 Perspectives on AI in Anaesthesiology
23.1 Introduction
23.2 AI in Peri-operative Patient Risk Stratification
23.3 AI in intra-operative Management
23.4 AI in Post-operative Management and Discharge Planning
23.5 Imaging and Technical Skills Aid
23.6 Conclusion
Conflicts of Interest
References
Chapter 24 AI in Ear, Nose, and Throat
24.1 Introduction
24.2 Ear, Nose, and Throat
24.3 Ear
24.4 Nose
24.5 Throat
24.6 The Future of AI in ENT
References
Chapter 25 AI in Obstetrics and Gynaecology
25.1 Introduction
25.2 Reproductive Medicine
25.3 Early Pregnancy
25.4 Antenatal Care
25.5 Pregnancy Ultrasonography
25.6 Foetal Heart Rate Monitoring
25.7 Intrapartum Care
25.8 Postnatal Care
25.9 Menopause
25.10 Gynaeoncology
25.11 The Future of AI in Obstetrics and Gynaecology
Declarations of Interest
References
Chapter 26 AI in Ophthalmology
26.1 Introduction
26.2 Looking into the Eye: A Historical Perspective
26.3 Deep Learning and Ophthalmology
26.4 AI-Informed Diagnostics in the Eye Clinic
26.5 Assistive Technology Applications of AI in Ophthalmology
26.6 Caveats and Challenges
26.7 The Future of Artificial Intelligence in Ophthalmology
References
Chapter 27 AI in Orthopaedic Surgery
27.1 Activity Tracking and Digital Outcomes
27.2 Image Processing and Analysis
27.3 Clinical Outcome Prediction and Decision Support
27.4 Health Systems Efficiency and Optimization
Conflicts of Interests
References
Chapter 28 AI in Surgery
28.1 Introduction
28.2 Pre-operative Care
28.3 Intra-operative Care
28.4 Postoperative Care and Long-Term Management
28.5 Surgical Training
28.6 Challenges and Opportunities
28.7 Caveats for AI in Surgery
28.8 Conclusion
References
Chapter 29 AI in Urological Oncology: Prostate Cancer Diagnosis with Magnetic Resonance Imaging
29.1 Introduction
29.2 Deep Learning Algorithms for Prostate Cancer Diagnosis
29.3 Future of AI in Prostate Cancer Diagnosis
References
Diagnostic Specialties
Chapter 30 AI in Pathology
30.1 Digital Pathology: Beginning of a New Era
30.2 Successful Applications of Clinical Pathology AI: The Basics
30.3 AI for Streamlined Clinical Workflow
30.4 Major Challenges
30.5 Future Potential
30.6 Conclusions
References
Chapter 31 Introduction to AI in Radiology
31.1 Introduction
31.2 AI in Diagnostic Imaging
31.3 Enhancement of Radiology Workflow
31.4 Medical Imaging Processing – Powered by AI
31.5 Future Outlook and Obstacles
References
Chapter 32 Clinical Applications of AI in Diagnostic Imaging
32.1 Introduction
32.2 Clinical AI Solutions in Radiology
32.3 Evaluation of AI Tools for Clinical Implementation
32.4 The Hidden Costs of Implementing AI in Practice
32.5 Conclusion
References
Chapter 33 AI for Workflow Enhancement in Radiology
33.1 Introduction to Imaging Workflow
33.2 Image Display Protocols
33.3 Quality Assurance and Peer Review: ‘Second Reader’ Applications
33.4 Clinical Decision Support
33.5 Natural Language Processing
33.6 Medical Image Protocolling
33.7 Image Acquisition and AI support
33.8 Imaging Pathway in an Emergency and Trauma Radiology Department
33.9 Challenges Related to AI Use
33.10 Conclusion
References
Chapter 34 AI for Medical Image Processing: Improving Quality, Accessibility, and Safety
34.1 Introduction
34.2 Basic Introduction to Medical Image Processing Concepts
34.3 Basic AI-Based Image Quality Improvement Approaches
34.4 Specific Problems and AI-Based Solutions
34.5 Potential Pitfalls
34.6 Looking Ahead: Beyond Image Acquisition
34.7 Conclusions
References
Chapter 35 Future Developments and Assimilation of AI in Radiology
35.1 Introduction
35.2 Future of Radiology Reporting
35.3 Beyond the Radiologist
35.4 Future for Patients
35.5 Future for Administrative Staff
35.6 Ethical Considerations in Radiology
35.7 Legal and Ethical Challenges in Radiology
35.8 Differential Role of AI across Imaging Modalities
35.9 Using Future Predictions to Guide Current Planning
35.10 Determining Who Pays
35.11 Conclusion
References
SECTION IV Policy Issues, Practical Implementation, and Future Perspectives in Medical AI
AI Regulation, Privacy, Law
Chapter 36 Medical Device AI Regulatory Expectations
36.1 Introduction
36.2 EU Regulatory Requirements for Medical Devices, Including Software and AI
36.3 Health Canada Regulatory Requirements for Medical Devices Including Software
36.4 FDA Regulatory Requirements for Medical Devices Including Software
36.5 FDA’s Action Plans for AI/Machine Learning in SaMD
36.6 Conclusion
References
Chapter 37 Privacy Laws in the USA, Europe, and South Africa
37.1 Introduction
37.2 United States of America
37.3 Europe
37.4 South Africa
37.5 Similarities and Differences between HIPAA, GDPR, and the POPI Act
37.6 Conclusion
Acknowledgments
References
Chapter 38 AI-Enabled Consumer-Facing Health Technology
38.1 Introduction
38.2 Trends and Patient Receptivity to AI-Enabled Consumer-Facing Health Technology
38.3 Different Types of AI-Enabled Consumer-Facing Health Technology
38.4 Common Issues for Consumer-Facing AI-Enabled Health Technology
38.5 Emerging Standards and Regulation Efforts
38.6 Best Practices for Physician Involvement with the Creation and Validation of Consumer-Facing AI-Enabled Health Technology
38.7 Best Practices for Recommending and Responding to Patient Queries about AI-Enabled Consumer-Facing Health Technology
38.8 Future Directions for AI-Enabled Consumer-Facing Health Technology
Conflicts of Interest
References
Ethics, Equity, Bias
Chapter 39 Biases in Machine Learning in Healthcare
39.1 Introduction
39.2 Pre-existing Disparities
39.3 Risks in Implementing Machine Learning in Healthcare
39.4 Next Steps and Solutions
References
Chapter 40 ‘Designing’ Ethics into AI: Ensuring Equality, Equity, and Accessibility
40.1 Introduction
40.2 Business and Use Case Development
40.3 Design Phase
40.4 Data
40.5 Building
40.6 Testing
40.7 Deployment
40.8 Monitoring
40.9 Conclusion
References
Design and Implementation
Chapter 41 Making AI Work: Designing and Evaluating AI Systems in Healthcare
41.1 Introduction
41.2 Background in Human–Computer Interaction
41.3 Designing Human-AI in Healthcare
41.4 Evaluating Human-AI in Healthcare
41.5 Open Research Questions
41.6 Conclusion
References
Chapter 42 Demonstrating Clinical Impact for AI Interventions: Importance of Robust Evaluation and Standardized Reporting
42.1 Clinical Evaluation and Study Design for AI in Clinical Medicine
42.2 Complete and Transparent Reporting of AI Studies
42.3 Conclusion
References
Chapter 43 The Importance and Benefits of Implementing Modern Data Infrastructure for Video-Based Medicine
43.1 Prevalence of Video in Medicine
43.2 The Solution: Thoughtful Video Data Capture
43.3 Options for Building Video Data Capture Infrastructure
43.4 How to Leverage Video Infrastructure for AI Development
43.5 IT, Security, and Legal Considerations When Implementing Video Capture Infrastructure
43.6 Ancillary Benefits of Implementing Video Capture in Clinical Practice
43.7 Conclusion
References
The Way Forward
Chapter 44 AI and the Evolution of the Patient–Physician Relationship
44.1 Introduction
44.2 Patient–Physician Relationship
44.3 AI in the Clinical Context
44.4 AI Impact Scenarios
44.5 AI Development with Humans
44.6 Conclusion
References
Chapter 45 Virtual Care and AI: The Whole Is Greater Than the Sum of Its Parts
45.1 Understanding Digital Transformation
45.2 Vision of Virtual Care and AI Synergy
45.3 Understanding the Virtual Care as a Service (VCaaS) Model
45.4 Understanding the AI as a Service (AIaaS) Model
45.5 Convergence and Synergy between VCaaS and AIaaS and Other Technologies
45.6 Evolution of Legal and Regulatory Frameworks
45.7 FDA Encourages the Use of AI in Virtual Care
45.8 Value and Reimbursement
45.9 Evolution of Stroke Care in the Digital Era
45.10 Full-Stack Digital Clinician
45.11 Conclusion
References
Chapter 46 Summing It All Up: Evaluation, Integration, and Future Directions for AI in Clinical Medicine
46.1 Introduction
46.2 Foundations of AI and Machine Learning
46.3 Types of Learning
46.4 AI Model Review
46.5 Data Quality and Data Standards
46.6 Measures of AI Model Performance
46.7 Validation of AI Systems
46.8 Review of AI Applications in Clinical Medicine
46.9 Future Directions
Conflicts of Interest
References
Chapter 47 A Glimpse into the Future: AI, Digital Humans, and the Metaverse – Opportunities and Challenges for Life Sciences in Immersive Ecologies
47.1 Introduction to the Metaverse
47.2 Digital Humans and Genomic Information
47.3 Immortality in the Metaverse
47.4 AI and VR in Drug Discovery
47.5 Immersive Environments in Life Sciences
47.6 Digital Twins
47.7 Gamification to Connect and Bring Healthcare Providers and Consumers Together
47.8 Telemedicine
47.9 Facilitating Collaboration Among Healthcare Professionals
47.10 Conclusions
References
Index
EULA