This book offers a comprehensive yet concise overview of the challenges and opportunities presented by the use of artificial intelligence in healthcare. It does so by approaching the topic from multiple perspectives, e.g. the nursing, consumer, medical practitioner, healthcare manager, and data analyst perspective. It covers human factors research, discusses patient safety issues, and addresses ethical challenges, as well as important policy issues. By reporting on cutting-edge research and hands-on experience, the book offers an insightful reference guide for health information technology professionals, healthcare managers, healthcare practitioners, and patients alike, aiding them in their decision-making processes. It will also benefit students and researchers whose work involves artificial intelligence-related research issues in healthcare.
Author(s): Mowafa Househ, Elizabeth Borycki, Andre Kushniruk
Series: Lecture Notes in Bioengineering
Publisher: Springer
Year: 2021
Language: English
Pages: 206
City: Cham
Preface
Acknowledgements
Contents
AI From Human, Management and Policy Perspectives
The Human Factors of AI in Healthcare: Recurrent Issues, Future Challenges and Ways Forward
1 Introduction
2 A Brief History of Human Factors Issues in Healthcare AI
3 Human Factors Issues and Challenges in Healthcare AI
4 Characteristics of Successful AI Applications in Healthcare
5 Discussion—Future Directions for Work in the Human Factors of AI
References
The Safety of AI in Healthcare: Emerging Issues and Considerations for Healthcare
1 Introduction
2 What is AI?
3 Technology-Induced Errors in Healthcare and AI Safety
4 Data and Safety of AI Solutions
5 Data, Safety and the Role of Large Datasets
6 Data, Safety, and Patient Privacy
7 Effects of Dataset Quality in Healthcare and Safety
8 AI in Healthcare Contexts
9 Research Evidence: How Do We Evaluate the Safety and Quality of AI?
10 AI and Its Implementation
11 Future Directions: AI and Patient Safety
12 Conclusion
References
Utilizing Health Analytics in Improving the Performance of Hospitals and Healthcare Services: Promises and Challenges
1 Introduction
1.1 Background
1.2 What Does Health Analytics do?
1.3 Types of Health Analytics
2 Methods
3 Results
3.1 Improving Healthcare Performance
3.2 Challenges of Utilizing Health Analytics
4 Discussion
4.1 Clinicians and Healthcare Professionals
4.2 Hospitals, Insurance, Pharmaceutical and Other Companies
4.3 Healthcare Government Organizations
5 Conclusions
References
Perspectives on Human-AI Interaction Applied to Health and Wellness Management: Between Milestones and Hurdles
1 Introduction
2 Milestones
2.1 Understandable, and Explainable AI
2.2 Documentation as an Integral Part of the Development Process
2.3 Incorporating both Artificial Intelligence and Human Intelligence
2.4 The Birth of Human-AI Interaction
3 Hurdles
3.1 Artificial Intelligence Literacy
3.2 Opaque Nature of Machine Learning Algorithms
3.3 The Design over Data Versus Data over Design Paradigm Dilemma
3.4 Control over Customized Functionalities for Niche User Segments
3.5 Foreseeing the Unforeseeable—Adapting to an Ever Changing Human-AI Interaction
4 Conclusion
References
Artificial Intelligence in Healthcare from a Policy Perspective
1 Introduction
2 Artificial Intelligence Policy
3 Accuracy, Fairness and Transparency
4 Data Privacy and Consent
5 Accountability
6 Workforce Disruption
7 Conclusions
References
Privacy-Preserving AI in Healthcare
1 Introduction
2 AI in Healthcare
3 Preserving Privacy
3.1 Homomorphic Encryption
3.2 Multi-party Computation
3.3 Differential Privacy
3.4 Federated Learning
4 Legal Framework
5 Concluding Remarks
References
Patients Perspective—Benefits and Challenges of Artificial Intelligence
1 Artificial Intelligence: A Toolbox of Potential
2 Artificial Intelligence and Decision Making in Health Systems
3 Ethical Concerns Around Use of Artificial Intelligence
4 Patient Perspectives About AI and Associated Health Technologies
5 Taiwan’s Health Technology Journey and Initiatives in Global Crisis
6 MyHealthbank—eMask Initiative
7 Conclusion
References
AI From a Health Professional Perspective
Artificial Intelligence and Medication Management
1 The Potential of Machine Learning Techniques for Medication Management
2 Challenges for the Development of Predictive Analytic Algorithms Related to Medication: Meaningful and Comprehensive Data
2.1 Estimating Medication Exposure
2.2 Estimating Positive and Negative Outcomes in a Computable and Coherent Way
3 Conclusion
References
Reflections on the Clinical Acceptance of Artificial Intelligence
1 Introduction
1.1 AI Nomenclature
2 AI in Clinical Environments
3 Challenges and Opportunities
3.1 Data Repositories
3.2 Clean Annotation and Labels
3.3 Local and Cloud-Based Learning
3.4 Model (Trained AI)
3.5 Clinical Recommendation System
3.6 Clinical Studies
3.7 Clinical Certification and Approval
3.8 Clinical Acceptance
3.9 Liability Risks
4 Conclusion
References
Artificial Intelligence for Chatbots in Mental Health: Opportunities and Challenges
1 Introduction
2 Chatbots for Mental Health
2.1 Overview of Chatbots for Mental Health
2.2 Role of AI in Mental Health Chatbots
3 Benefits of Chatbots in Mental Health
4 Challenges
4.1 Technical Limitations of Mental Health Chatbots
4.2 Ethical Challenges
4.3 Impact on User and Healthcare Team
4.4 Accountability Implications
5 Future Research Directions
References
AI and Machine Learning in Diabetes Management: Opportunity, Status, and Challenges
1 What is Diabetes?
1.1 Forms of Diabetes
1.2 Diagnosis
2 Importance of Diabetes Management
2.1 Retinopathy and Blindness
2.2 Kidney Disease
2.3 Heart Disease and High Blood Pressure
2.4 Other Diabetes Associated Diseases
2.5 Effective Management Technologies
3 The Integration of AI and Machine Learning for Diabetes Care
3.1 Estimated HbA1C Versus Predictive HbA1C
3.2 CGM Based Hyper and Hypoglycemia Predictions
4 Long Term Unmet Challenges
5 Future Work
References
AI From a Technological Perspective
Reinforcement Learning Applications in Health Informatics
1 Introduction
2 Reinforcement Learning
3 Reinforcement Learning in Healthcare
3.1 Healthcare IoT
3.2 Medication Dosing
3.3 Drug Design
3.4 Treatment Recommendation
3.5 Lung Radiotherapy
3.6 Personal Health Advisor
3.7 Sepsis Treatment
4 Summary
5 Conclusion
References
Deep Learning in Healthcare
1 Deep Learning—An Overview
2 Scope of Deep Learning in Healthcare
2.1 Cancer Diagnosis
3 Deep Learning and Cancer
3.1 The Threat of Cancer
3.2 Cancer Diagnosis and Deep Learning
3.3 Deep Learning in Lung Cancer Detection
3.4 Google Research on Lung Cancer Detection
3.5 Detecting Cancer with Double Convolutional Network
3.6 Deep Learning in Breast Cancer Detection
3.7 Deep Learning in Brain Cancer Detection
4 Scope Versus Challenges in Deep Learning for Healthcare
5 Conclusion
Bibliography
Deep Learning in Biomedical Text Mining: Contributions and Challenges
1 Introduction
2 Deep Learning Architectures and Techniques that Have Been Proven Successful in NLP
2.1 Embeddings
2.2 Classical DL Based Techniques: CNN, RNN, LSTM, Attention Mechanism
2.3 Transfer Learning and Recent DL-Based Architectures that Rejuvenate the NLP Domain
3 Deep Learning for Named-Entity Recognition in Biomedical Texts
4 Deep Learning for Relationship Extraction from Biomedical Texts
5 Deep Learning for Question Answering from Biomedical Texts
6 Challenges and Future Perspectives
7 Conclusions
References
Artificial Intelligence in the Fight Against the COVID-19 Pandemic: Opportunities and Challenges
1 Introduction
2 AI Potentials in the Fight Against COVID-19
2.1 Early Warnings
2.2 Forecasting the Epidemic Development
2.3 Early Detection and Diagnosis
2.4 Prognosis Prediction
2.5 Treatments and Vaccine Development
2.6 Social Control
2.7 Infodemiology
2.8 Other opportunities
3 Challenges in Leveraging AI in the Fight Against COVID-19
3.1 Data Challenges
3.2 Maturity and Acceptance Challenges
3.3 Ethical Challenges
3.4 Privacy Challenges
3.5 Explainable AI
4 Conclusion
References
