Artificial Intelligence in Healthcare

Artificial Intelligence in Healthcare
Copyright
Endorsement
Contents
List of contributors
About the editors
Biographies
Preface
	About this book
	Intended audience
	How is this book organized
Introduction
	The promise of an intelligent machine
	Current applications and challenges in healthcare
1 Current healthcare, big data, and machine learning
	1.1 Current healthcare practice
		1.1.1 The rising need for technology
		1.1.2 New models in healthcare
	1.2 Value-based treatments and healthcare services
		1.2.1 Value-based healthcare
		1.2.2 Increasing health outcomes
		1.2.3 Patient-centered care (the patient will see you now)
		1.2.4 Personalized medicine
	1.3 Increasing data volumes in healthcare
		1.3.1 Big data and data accumulation
		1.3.2 Data generation sources
		1.3.3 Big data types
	1.4 Analytics of healthcare data (machine learning and deep learning)
		1.4.1 Machine learning
		1.4.2 Deep learning
	1.5 Conclusions/summary
	References
2 The rise of artificial intelligence in healthcare applications
	2.1 The new age of healthcare
		2.1.1 Technological advancements
		2.1.2 Artificial intelligence applications in healthcare
	2.2 Precision medicine
		2.2.1 Genetics-based solutions
		2.2.2 Drug discovery and development
			2.2.2.1 Drug property and activity prediction
			2.2.2.2 De novo design through deep learning
			2.2.2.3 Drug–target interactions
	2.3 Artificial intelligence and medical visualization
		2.3.1 Machine vision for diagnosis and surgery
			2.3.1.1 Computer vision for diagnosis and surgery
		2.3.2 Deep learning and medical image recognition
		2.3.3 Augmented reality and virtual reality in the healthcare space
			2.3.3.1 Education and exploration
			2.3.3.2 Patient experience
	2.4 Intelligent personal health records
		2.4.1 Health monitoring and wearables
		2.4.2 Natural language processing
		2.4.3 Integration of personal records
	2.5 Robotics and artificial intelligence-powered devices
		2.5.1 Minimally invasive surgery
		2.5.2 Neuroprosthetics
	2.6 Ambient assisted living
		2.6.1 Smart home
		2.6.2 Assistive robots
		2.6.3 Cognitive assistants
		2.6.4 Social and emotional stimulation
	2.7 The artificial intelligence can see you now
		2.7.1 Artificial intelligence in the near and the remote
		2.7.2 Success factors for artificial intelligence in healthcare
			2.7.2.1 Assessment of condition
			2.7.2.2 Managing complications
			2.7.2.3 Patient-care assistance
			2.7.2.4 Medical research
		2.7.3 The digital primary physician
			2.7.3.1 Artificial intelligence prequalification (triage)
			2.7.3.2 Remote digital visits
			2.7.3.3 The future of primary care
	References
3 Drug discovery and molecular modeling using artificial intelligence
	3.1 Introduction. The scope of artificial intelligence in drug discovery
		3.1.1 Areas in which machine learning techniques are applied in biotechnology
	3.2 Various types of machine learning in artificial intelligence
		3.2.1 Artificial neural networks as tools in drug discovery
		3.2.2 Architecture of artificial neural network for drug discovery applications
		3.2.3 Artificial neural network methods for structure prediction in proteins from their sequence
		3.2.4 Artificial neural network methods for spectroscopy in biomedicine
	3.3 Molecular modeling and databases in artificial intelligence for drug molecules
		3.3.1 Databases for the training sets in drug discovery
		3.3.2 Database mining
	3.4 Computational mechanics ML methods in molecular modeling
		3.4.1 Molecular dynamics simulation for drug development
		3.4.2 Computations with quantum mechanical techniques for drug development
	3.5 Drug characterization using isopotential surfaces
	3.6 Drug design for neuroreceptors using artificial neural network techniques
	3.7 Specific use of deep learning in drug design
		3.7.1 Other applications of machine learning in drug development
	3.8 Possible future artificial intelligence development in drug design and development
	References
4 Applications of artificial intelligence in drug delivery and pharmaceutical development
	4.1 The evolving pharmaceutical field
	4.2 Drug delivery and nanotechnology
	4.3 Quality-by-design R&D
	4.4 Artificial intelligence in drug delivery modeling
	4.5 Artificial intelligence application in pharmaceutical product R&D
		4.5.1 Artificial intelligence application in prototyping and early development: an example scenario
		4.5.2 Artificial intelligence in late-stage development: an example scenario
			4.5.2.1 Artificial intelligence in manufacturing development and control
	4.6 Landscape of AI implementation in the drug delivery industry
	4.7 Conclusion: the way forward
	References
5 Cancer diagnostics and treatment decisions using artificial intelligence
	5.1 Background
	5.2 Artificial intelligence, machine learning, and deep learning in cancer
	5.3 Artificial intelligence to determine cancer susceptibility
	5.4 Artificial intelligence for enhanced cancer diagnosis and staging
	5.5 Artificial intelligence to predict cancer treatment response
	5.6 Artificial intelligence to predict cancer recurrence and survival
	5.7 Artificial intelligence for personalized cancer pharmacotherapy
	5.8 How will artificial intelligence affect ethical practices and patients?
	5.9 Concluding remarks
	References
6 Artificial intelligence for medical imaging
	6.1 Introduction
	6.2 Outputs of artificial intelligence in radiology/medical imaging
		6.2.1 Preprocessing
		6.2.2 Segmentation
		6.2.3 Object detection
	6.3 Using artificial intelligence in radiology and overcoming its hurdles
		6.3.1 Not enough training data (data augmentation)
		6.3.2 Unbalanced training data (data weighing)
		6.3.3 Not representative training data (transfer learning, domain adaptation)
			6.3.3.1 Transfer learning
			6.3.3.2 Domain adaptation
				6.3.3.2.1 Black box—algorithm explanation
				6.3.3.2.2 Implementation/integration
	6.4 X-rays and artificial intelligence in medical imaging—case 1 (Zebra medical vision)
		6.4.1 X-rays and their role in medicine
		6.4.2 X-ray discovery
		6.4.3 Chest X-rays
		6.4.4 RadBot-CXR—clinical findings using deep learning
		6.4.5 Detecting osteoporosis using artificial intelligence
	6.5 Ultrasound and artificial intelligence in medical imaging—case 2 (Butterfly iQ)
		6.5.1 Ultrasound and its role in medicine
		6.5.2 The Butterfly iQ
	6.6 Application of artificial intelligence in medical imaging—case 3 (Arterys)
	6.7 Perspectives
	References
7 Medical devices and artificial intelligence
	7.1 Introduction
	7.2 The development of artificial intelligence in medical devices
		7.2.1 Activity tracking devices
		7.2.2 Implants, bionics and robotics devices
	7.3 Limitations of artificial intelligence in medical devices
	7.4 The future frontiers of artificial intelligence in medical devices
	References
8 Artificial intelligence assisted surgery
	8.1 Introduction
	8.2 Preoperative
		8.2.1 Preoperative risk assessment
		8.2.2 Preoperative diagnosis
		8.2.3 Preoperative staging
	8.3 Intraoperative
		8.3.1 Autonomous surgery
		8.3.2 Computer vision
	8.4 Postoperative
		8.4.1 Detecting complications
		8.4.2 Training and certification of surgeons
	8.5 Conclusion
	References
	Further reading
9 Remote patient monitoring using artificial intelligence
	9.1 Introduction to remote patient monitoring
	9.2 Deploying patient monitoring
		9.2.1 Patient monitoring in healthcare today
		9.2.2 Using remote patient monitoring to improve clinical outcomes
	9.3 The role of artificial intelligence in remote patient monitoring
		9.3.1 Harnessing the power of consumer technology
		9.3.2 Sensors, smartphones, apps, and devices
		9.3.3 Natural language processing
		9.3.4 Natural language processing technology applications in healthcare
		9.3.5 Clinical decision support
		9.3.6 Ambient assisted living
	9.4 Diabetes prediction and monitoring using artificial intelligence
		9.4.1 DiaBits
		9.4.2 Other diabetes remote monitoring apps and devices
	9.5 Cardiac monitoring using artificial intelligence
		9.5.1 Virtual application of artificial intelligence in cardiology
			9.5.1.1 Imaging interpretation
			9.5.1.2 Clinical decision support systems
			9.5.1.3 Artificial intelligence in virtual reality, augmented reality, and voice powered virtual assistants
			9.5.1.4 “BIG DATA” for predictive analysis
	9.6 Neural applications of artificial intelligence and remote patient monitoring
		9.6.1 Dementia
			9.6.1.1 Artificial intelligence in dementia monitoring
			9.6.1.2 Smart homes to support dementia patients
		9.6.2 Migraine
			9.6.2.1 Migraine Buddy
			9.6.2.2 Manage My Pain Pro
	9.7 Conclusions
	References
10 Security, privacy, and information-sharing aspects of healthcare artificial intelligence
	10.1 Introduction to digital security and privacy
	10.2 Security and privacy concerns in healthcare artificial intelligence
		10.2.1 Defining privacy
		10.2.2 Privacy and data sharing
			10.2.2.1 Toward data interoperability in healthcare
		10.2.3 Safety and security in an era of emerging technologies in healthcare
		10.2.4 Measures to protect sensitive data in healthcare
			10.2.4.1 Traditional techniques
			10.2.4.2 Emerging techniques
	10.3 Artificial intelligence’s risks and opportunities for data privacy
		10.3.1 Artificial intelligence’s risks for data privacy
			10.3.1.1 Susceptibility to data privacy and integrity attacks
				10.3.1.1.1 Reidentification of patients using artificial intelligence
				10.3.1.1.2 Data integrity and bias
				10.3.1.1.3 Inadvertent disclosure
			10.3.1.2 More limited level of control over data ownership and handling
			10.3.1.3 Less intuitive understanding of one’s data privacy
			10.3.1.4 Use of sensitive health data outside of healthcare
			10.3.1.5 Privacy externalities
		10.3.2 Artificial intelligence’s opportunities for data privacy
			10.3.2.1 Improving data utility
			10.3.2.2 Better attribution of privacy violations
			10.3.2.3 Improving physician productivity and data sharing
	10.4 Addressing threats to health systems and data in the artificial intelligence age
		10.4.1 Addressing cyberattacks and data breaches in healthcare
		10.4.2 Government regulation of security and privacy in healthcare
	10.5 Defining optimal responses to security, privacy, and information-sharing challenges in healthcare artificial intelligence
		10.5.1 Governments
			10.5.1.1 Regulation
			10.5.1.2 Oversight and enforcement
			10.5.1.3 National strategy development
		10.5.2 Innovators
			10.5.2.1 Build sustainable solutions
			10.5.2.2 Adopting privacy principles
		10.5.3 Developing privacy-centered solutions
		10.5.4 Providers
			10.5.4.1 Internal review
			10.5.4.2 Governance
			10.5.4.3 Education and training
			10.5.4.4 Communication
		10.5.5 Patients
			10.5.5.1 Data ownership
			10.5.5.2 Consent granting
	10.6 Conclusions
	Acknowledgements
	References
11 The impact of artificial intelligence on healthcare insurances
	11.1 Overview of the global health insurance industry
	11.2 Key challenges facing the health insurance industry
		11.2.1 Cost of healthcare
		11.2.2 Increased expectations of customers
		11.2.3 Regulatory overhead
	11.3 The application of artificial intelligence in the health insurance industry
		11.3.1 Overview and history of artificial intelligence
		11.3.2 Artificial intelligence in health insurance
			11.3.2.1 Improving customer experience
			11.3.2.2 Reducing fraud
			11.3.2.3 Improving back-office efficiency and reducing cost
			11.3.2.4 Reducing risk and optimizing premiums
			11.3.2.5 Creating new business opportunities
	11.4 Case studies
		11.4.1 Successes
		11.4.2 Failures
	11.5 Moral, ethical, and regulatory concerns regarding the use of artificial intelligence
		11.5.1 Regulatory challenges
		11.5.2 Ethical challenges
	11.6 The limitations of artificial intelligence
		11.6.1 Lack of talent
		11.6.2 Lack of data or poor data quality
		11.6.3 Artificial intelligence accuracy, bias, and relevance
		11.6.4 Artificial intelligence as part of the decision process
		11.6.5 Process control
	11.7 The future of artificial intelligence in the health insurance industry
	References
12 Ethical and legal challenges of artificial intelligence-driven healthcare
	12.1 Understanding “artificial intelligence”
	12.2 Trends and strategies
		12.2.1 United States
		12.2.2 Europe
	12.3 Ethical challenges
		12.3.1 Informed consent to use
		12.3.2 Safety and transparency
		12.3.3 Algorithmic fairness and biases
		12.3.4 Data privacy
	12.4 Legal challenges
		12.4.1 Safety and effectiveness
			12.4.1.1 United States
				12.4.1.1.1 Medical devices
				12.4.1.1.2 Medical and certain decision support software
					12.4.1.1.2.1 Software functions under Section 520(o)(1)(A)–(D) of the FDCA
					12.4.1.1.2.2 Software functions under Section 520(o)(1)(E) of the FDCA
				12.4.1.1.3 Other FDA initiatives
			12.4.1.2 Europe
				12.4.1.2.1 Medical devices and new legal developments
				12.4.1.2.2 MDR
		12.4.2 Liability
			12.4.2.1 United States
			12.4.2.2 Europe
		12.4.3 Data protection and privacy
			12.4.3.1 United States
			12.4.3.2 Europe
		12.4.4 Cybersecurity
		12.4.5 Intellectual property law
	12.5 Conclusion
	Acknowledgements
	References
Concluding remarks
Index