Recent advances in artificial intelligence (AI) and machine learning have witnessed many successes in various disciplines including the healthcare sector. Innovations in intelligent medical systems have revolutionized the way in which healthcare services are provided, ranging from making clinical diagnosis, developing personalized treatment and drugs, assisting patient monitoring, to automating administrative tasks and reducing operational costs. In this book, the authors present key applications in the general area of health care, where AI has made significant successes. From the individual chapters, the readers will be provided with a range of examples to illustrate the wide plethora of application domains utilizing state-of-the-art AI techniques, proving credence to the versatility and effectiveness of an AI approach in health care and medicine. We envisage that this book is ideal for individuals new to the notion of AI in health care, equally, early career academics who wish to further expand on their knowledge in AI in medicine. What will be presented is in no means an exhaustive list of applications, but most definitely a varied one.
Author(s): Tianhua Chen, Jenny Carter, Mufti Mahmud, Arjab Singh Khuman
Series: Brain Informatics and Health
Publisher: Springer
Year: 2022
Language: English
Pages: 315
City: Singapore
Preface
Contents
AI in Healthcare: Malignant or Benign?
1 Introduction
2 Prevention
2.1 AI as a Tool to Change the World
2.2 Governance, Laws and Ethics
2.3 Touchless Control Preventative Healthcare in COVID
2.4 Personalised Medicine: The Bridge Between Prevention and Diagnosis
3 Diagnosis
3.1 History
3.2 Fuzzy Diagnosis
3.3 NLP and Diagnostic Chatbots
3.4 Medical Imaging
3.5 Mental Health
3.6 Problems with AI Diagnosis
4 Care
4.1 The Problem with Current Care
4.2 Mental Health Chat Bots
4.3 Companionship for the Elderly
4.4 Sex Bots
4.5 Summary and Problems
5 Cure
5.1 Discovery
5.2 Prosthetics, Implants and Exoskeletons
6 Conclusion
References
Process Mining in Healthcare: Challenges and Promising Directions
1 Introduction
2 Process Mining for Healthcare
2.1 Process Discovery in Healthcare
2.2 Conformance Checking in Healthcare
2.3 Process Enhancement in Healthcare
3 Challenges and Promising Directions
3.1 Data and Data Sources
3.2 Scheduling and Planning
3.3 Patients' Privacy and Multicentric Clinical Studies
3.4 Explainability and Understandability
4 Conclusion
References
Computational Intelligence in Drug Discovery for Non-small Cell Lung Cancer
1 Introduction
1.1 In Silico Methods of Drug Discovery
2 Methodology and Experimentation
2.1 Data Collection
2.2 Data Cleaning
2.3 Feature Selection
2.4 Application of Machine Learning
2.5 Network Inference
2.6 Application of Deep Learning
2.7 KnowledgeFlow
3 Results and Discussion
3.1 Traditional Machine Learning for TCGA and GEO Data
3.2 Network Interactions
3.3 Performance with Deep Learning
4 Conclusion
References
AI for Brain Disorders
The Emerging Role of AI in Dementia Research and Healthcare
1 Introduction to Dementia Research and Healthcare
2 Genetics
2.1 How Do We Determine the Biological Effect of Genetic Variants?
2.2 How Do We Get from Genetic Epidemiology and -Omics to Practical Applications?
3 Experimental Medicine
3.1 What Makes a Good Experimental Model?
3.2 How Can We Make Best Use of Multimodal Data?
3.3 How Can We Translate Insights from Experimental Models to Human Disease Biology?
4 Drug Discovery and Trials Optimisation
5 Neuroimaging
6 Prevention
7 A Global Initiative for AI Applied to Dementia Research and Healthcare
8 Conclusions
References
Effective Diagnosis of Parkinson’s Disease Using Machine Learning Techniques
1 Introduction
2 Related Works
3 Experimental Pipeline and Setup
4 Experimental Results and Discussion
5 Conclusion
References
Brain Networks in Autism Spectrum Disorder, Epilepsy and Their Relationship: A Machine Learning Approach
1 Introduction
2 Mapping Eeg Time-Series to Complex Network
3 Material and Method
3.1 Dataset
3.2 Network Metrics and Analysis
3.3 Classification and Performance Evaluation of Metrics
4 Results
4.1 Brain Network Topology
4.2 Statistical Analysis of Network Metrics
4.3 Assessment With Standardized Data
5 Discussion
6 Conclusion
References
AI in Mental Health
Computational Intelligence in Depression Detection
1 Introduction
2 Literature Review
3 Depression Detection
4 Computational Intelligence in Depression Detection
4.1 Depression Detection from Social Media Data
4.2 Depression Detection from Image/Video Data
4.3 Depression Detection from Bio Signal
4.4 Depression Detection from Smartphone Data
5 Challenges and Research Direction
5.1 Gaps in the Literature
5.2 Future Research Scopes
6 Conclusion
References
Investigating Mental Wellbeing in the Technology Workplace Using Machine Learning Techniques
1 Introduction
2 Materials and Methods
2.1 The Data
2.2 Data Preprocessing
2.3 Flowchart
3 Experimentation and Discussion
3.1 Cluster Analysis
3.2 Visualisation
3.3 Building a Predictive Model Using Artificial Neural Networks
4 Conclusion
References
Computational Intelligence in Detection and Support of Autism Spectrum Disorder
1 Introduction
2 Computational Intelligence
2.1 Neural Networks
2.2 Fuzzy Logic
2.3 Evolutionary Computation
3 Computational Intelligence in Autism Detection
3.1 Datasets and Methods
4 Computational Intelligence in Autism Management
4.1 Apps and Platforms for Supporting People with Autism
5 Challenges and Future Research
6 Conclusion
References
AI for COVID-19
A Case Study of Using Machine Learning Techniques for COVID-19 Diagnosis
1 Introduction
2 Literature Review
3 Experimentation and Discussion
3.1 Data Preprocessing
3.2 Data Sampling
3.3 Model Selection
3.4 Hyperparameters Optimization
3.5 Evaluation
4 Conclusion
References
A Fuzzy Logic Approach to a Hybrid Lexicon-Based Sentiment Analysis Detection Tool Using Healthcare Covid-19 News Articles
1 Introduction
2 What is Sentiment Analysis?
3 Consensus Sentiment Analysis of News Articles Using a Fuzzy Inference System
4 Data Pre-processing and Sentiment Analysis
5 Data Transformation Process
6 Sentiment Analysis Algorithms
7 Fuzzification
8 Designing the Fuzzification Interface?
9 Functionality Block
10 Rule-Base
11 Membership Functions
12 Multi-inference System
13 Defuzzification
14 Cogs
15 Boa
16 Mom
17 Post Processing
18 Conclusion
References
AI for Cardivascular Diseases
Using Fuzzy Logic to Diagnose Blood Pressure
1 Introduction
2 Literature Review
3 System Overview
3.1 Approach to the Problem
3.2 System Description
4 System Testing
4.1 Data Collection for Testing
5 Critical Analysis
6 Conclusion
References
An AI-Based Approach to Identifying High Impact Comorbidities in Public Health Management of Diseases
1 Introduction
2 Background
2.1 Non-random Comorbidities
2.2 Coronary Heart Disease (CHD) in the UK
2.3 The “Small Changes” Philosophy
2.4 The Market Target (mt) Model
3 Case Study: Using the mt Model to Support Decision-Making in the Public Health Management of CHD in the UK
3.1 Overview
3.2 Results and Discussion
3.3 Deaths by CHD
3.4 Prevalence Versus Deaths
4 Concluding Remarks
4.1 Limitations
References
Risk Detection of Heart Disease
1 Introduction
2 Related Work
3 System Overview
4 System Design and Configuration
4.1 System Inputs
4.2 System Output
5 System Performance
6 Critical Evaluation
7 Conclusion
References
AI for Diabetes
A Case Study of Diabetes Diagnosis Using a Neuro-Fuzzy System
1 Introduction
2 Literature Review
3 Methods
4 Experimentation
5 Conclusion
References
A Fuzzy Logic Risk Assessment System for Type 2 Diabetes
1 Introduction
2 Related Research
2.1 What is Fuzzy Logic, and How Does It Work?
2.2 How Can Fuzzy Logic Be Applied to the Healthcare Field?
2.3 How Does Fuzzy Logic Improve This System?
3 Final System Overview
3.1 Description
3.2 Patient/User Risk Inference Subsystem
3.3 Doctor Risk Inference Subsystem (Optional)
3.4 Total Risk Inference Subsystem (Requires Other Subsystems)
4 Design Evolution and Evaluation
4.1 Initial System
4.2 Test Data and Data Generation
4.3 Test and Test Results Summary
4.4 Final System
5 Reflection
5.1 Testing
5.2 Effects of Different Risk Factors
5.3 System Performance and Possible Improvements
6 Conclusion
References
