This book reviews the application of artificial intelligence and machine learning in healthcare. It discusses integrating the principles of computer science, life science, and statistics incorporated into statistical models using existing data, discovering patterns in data to extract the information, and predicting the changes and diseases based on this data and models. The initial chapters of the book cover the practical applications of artificial intelligence for disease prognosis & management. Further, the role of artificial intelligence and machine learning is discussed with reference to specific diseases like diabetes mellitus, cancer, mycobacterium tuberculosis, and Covid-19. The chapters provide working examples on how different types of healthcare data can be used to develop models and predict diseases using machine learning and artificial intelligence. The book also touches upon precision medicine, personalized medicine, and transfer learning, with the real examples. Further, it also discusses the use of machine learning and artificial intelligence for visualization, prediction, detection, and diagnosis of Covid -19. This book is a valuable source of information for programmers, healthcare professionals, and researchers interested in understanding the applications of artificial intelligence and machine learning in healthcare.
Author(s): Ankur Saxena, Shivani Chandra
Publisher: Springer
Year: 2021
Language: English
Pages: 247
City: Singapore
Contents
About the Authors
List of Figures
List of Tables
1: Practical Applications of Artificial Intelligence for Disease Prognosis and Management
1.1 Overview of Application of AI in Disease Management
1.1.1 Disease Prognosis and Diagnosis
1.1.2 AI in Identification of Biomarker of Disease
1.1.3 AI in Drug Development
1.2 Public Data Repositories
1.2.1 KAGGLE
1.2.2 Csv
1.2.3 JSON
1.2.4 SQLite
1.2.5 Archives
1.2.6 UCI ML Repository
1.2.7 HealthData.gov
1.3 Review of Artificial Intelligence Techniques on Disease Data
1.3.1 Logistic Regression Model
1.3.2 Artificial Neural Network Model
1.3.3 Support Vector Machine Model
1.4 Case Study: Parkinson´s Disease Prediction
1.4.1 Importing the Data
1.4.2 Data Preprocessing and Feature Selection
1.4.3 Building Classifier
1.4.4 Predictive Modelling
1.4.5 Performance Validation of the Model
References
2: Automated Diagnosis of Diabetes Mellitus Based on Machine Learning
2.1 Introduction
2.2 Diabetes Mellitus
2.2.1 Classification of Diabetes Mellitus
2.2.2 Diagnosis of Diabetes Mellitus
2.2.3 Diabetes Management
2.3 Role of Artificial Intelligence in Healthcare
2.4 AI Technologies Accelerate Progress in Medical Diagnosis
2.5 Machine Learning
2.5.1 Types of Machine Learning
2.5.2 Role of Machine Learning in Diabetes Mellitus Management
2.6 Methodology for Development of an Application Based on ML
2.6.1 Dataset
2.6.2 Data Preprocessing
2.6.3 Model Construction
2.6.4 Results
2.7 Conclusion
References
3: Artificial Intelligence in Personalized Medicine
3.1 Introduction
3.2 Personalized Medicine
3.3 Importance of Artificial Intelligence
3.4 Use of Artificial Intelligence in Healthcare
3.5 Models of Artificial Intelligence Used in Personalized Medicine
3.6 Use of Different Learning Models in Personalized Medicine
3.6.1 Naïve Bayes Model
3.6.2 Support Vector Machine (SVM)
3.6.3 Deep Learning
References
4: Artificial Intelligence in Precision Medicine: A Perspective in Biomarker and Drug Discovery
4.1 Precision Medicine as a Process: A New Approach for Healthcare
4.2 Role of Artificial Intelligence: Biomarker Discovery for Precision Medicine
4.2.1 Biomarker(s) for Diagnostics
4.2.2 Biomarker(s) for Disease Prognosis
4.3 Role of Artificial Intelligence: Drug Discovery for Precision Medicine
4.3.1 Drug Discovery Process
4.3.2 Understanding the Disease Process and Target Identification
4.3.3 Identification of Hit and Lead
4.3.4 Synthesis of Compounds
4.3.5 Predicting the Drug-Target Interactions Using AI
4.3.6 Artificial Intelligence in Clinical Trials
4.3.7 Drug Repurposing
4.3.8 Some Examples of AI and Pharma Partnerships
4.4 Precision Medicine and Artificial Intelligence: Hopes and Challenges
References
5: Transfer Learning in Biological and Health Care
5.1 Introduction
5.2 Methodology
5.2.1 Dataset Curation
5.2.2 Data Loading and Preprocessing
5.2.3 Loading Transfer Learning Models
5.2.3.1 VGG-16
5.2.3.2 EfficientNet
5.2.3.3 Inception-ResNet-V2
5.2.3.4 Inception V3
5.2.4 Training
5.2.5 Testing
References
6: Visualization and Prediction of COVID-19 Using AI and ML
6.1 Introduction
6.2 Technology for ML and AI in SARS-CoV-2 Treatment
6.3 SARS-Cov-2 Tracing Using AI Technologies
6.4 Forecasting Disease Using ML and AI Technology
6.5 Technology of ML and AI in SARS-CoV-2 Medicines and Vaccine
6.6 Analysis and Forecasting
6.6.1 Predictions on the First Round
6.6.2 Predictions on the Second Round
6.6.3 Predictions on the Third Round
6.6.4 Predictions on the Fourth Round
6.6.5 Predictions on the Fifth Round
6.7 Methods Used in Predicting COVID-19
6.7.1 Recurrent Neural Networks (RNN)
6.7.2 Long Short-Term Memory (LSTM) and Its Variants
6.7.3 Deep LSTM/Stacked LSTM
6.7.4 Bidirectional LSTM (Bi-LSTM)
6.8 Conclusion
References
7: Machine Learning Approaches in Detection and Diagnosis of COVID-19
7.1 Introduction
7.2 Review of ML Approaches in Detection of Pneumonia in General
7.3 Application of Deep Learning Approaches in COVID-19 Detection
7.3.1 Deep Learning Model Frameworks
7.3.1.1 ResNet Models
7.3.1.2 Other CNN Models
7.3.2 The Data Imbalance Challenge
7.3.3 Interpretation/Visualization of Results
7.3.4 Performance Measurement Metrics
7.4 Challenges
7.5 Summary
References
8: Applications of Machine Learning Algorithms in Cancer Diagnosis
8.1 Introduction
8.1.1 Machine Learning in Healthcare
8.1.2 Cancer Study Using ML
8.2 Machine Learning Techniques
8.3 Machine Learning and Cancer Prediction/Prognosis
8.3.1 Cancer: The Dreaded Disease and a Case Study for ML
8.3.2 Machine Learning in Cancer
8.3.3 Dataset for Cancer Study
8.3.4 Steps to Implement Machine Learning
8.3.5 Tool Selection for Cancer Predictions
8.3.6 Methodology, Selection of ML Algorithm, and Metrics for Performance Measurement of ML in Cancer Prognosis
8.4 Results and Analysis
8.4.1 Liver Cancer Dataset
8.4.2 Prostate Cancer Dataset
8.4.3 Breast Cancer Dataset
8.5 Major Findings and Issues
8.6 Future Possibilities and Challenges in Cancer Prognosis
References
9: Use of Artificial Intelligence in Research and Clinical Decision Making for Combating Mycobacterial Diseases
9.1 Introduction of Technological Advancements and High Throughput Data in Genomics and Proteomics Work
9.1.1 High Throughput Screening of Tuberculosis
9.1.2 High Throughput Screening of Leprosy
9.1.3 High Throughput and Ultra-High Throughput Screening of Compound Libraries for Drug Discovery and Drug Repurposing
9.2 High Volume Data and the Bottleneck in Data Analysis
9.2.1 Development of Omics Data
9.2.2 NGS and its Use in Clinical Decision-Making, Proteomics, Docking, Simulations, Drug Screening (Repurposing of Drugs)
9.3 Advent of Artificial Intelligence (AI) and Machine Learning (ML)
9.3.1 Machine Learning and Deep Learning (DL) Algorithms
9.3.2 AI in Drug Repurposing
9.3.3 Examples from NGS and its Use in Clinical Decision-Making, Proteomics, Docking, Simulations, Drug Screening (Repurposing...
9.4 Illustrations of Machine Learning in Different Research Fields
9.4.1 AI and ML in Covid-19-Related Research
9.4.2 AI and ML in Skin Diseases
9.5 Limitations of AI and ML
9.6 Can Machines Become a Total Replacement for Human Intelligence?
9.7 Concluding Remarks
References
10: Bias in Medical Big Data and Machine Learning Algorithms
10.1 Introduction
10.2 Medical Big Data (MBD)
10.3 Analysis of Medical Big Data
10.4 Bias
10.4.1 Perceptive Bias
10.4.1.1 Problem Definition
10.4.1.2 Social and Technical Aspects
10.4.1.3 Fairness of Data
10.4.2 Processing Bias
10.4.2.1 Pre-Processing
10.4.2.2 In-Processing
10.4.2.3 Post Processing
10.4.3 Computing Bias
10.4.3.1 Awareness of Bias
10.4.3.2 Modelling Bias
10.4.3.3 AI Decisions
10.5 Conclusion
References
