Preface xiii List of Figures xv List of Tables xix List of Contributors xxi List of Notations and Abbreviations xxv 1 Importance of Analyzing Causality for Diabetes Care 1 1.1 Prevalence of Diabetes 2 1.2 Factors Contributing to Diabetes 5 1.3 Decision Support Tools and Diabetes Management 6 1.4 Conclusions 9 2 Advances and Opportunities in Digital Diabetic Healthcare Systems 15 2.1 Background 16 2.2 Transformation of Diabetes Management from Conventional to Digital 18 2.3 Digital Technologies for Diabetes Management 20 2.3.1 Artificial intelligence and machine learning (AI and ML) 21 2.3.2 Medical and healthcare Internet of Things (MHIoT) 25 2.3.3 Blockchain 27 2.3.4 Telemedicine 30 2.3.4.1 Telephonic/mobile care 31 2.3.4.2 Systems for continuous glucose monitoring 33 2.3.4.3 Diabetic retinopathy control techniques 33 2.3.4.4 Insulin pen 34 2.3.4.5 Insulin therapy decision support systems 34 2.3.4.6 Asymptomatic diabetes screening 34 2.3.5 mHealth 36 2.4 Current Challenges and Future Perspective 41 2.5 Conclusion 45 3 Role of IoT and Expert System in Diabetes Control with Continuous Diagnosis of Medical Conditions 53 3.1 Introduction 54 3.2 History of Diabetes 55 3.3 Diabetes 55 3.3.1 Type-1 diabetes 56 3.3.2 Type 2 diabetes 56 3.3.3 Gestational diabetes 57 3.3.4 Pre-Diabetes 57 3.4 Progressive Nature of Diabetes 57 3.5 Symptoms of Diabetes 58 3.6 Individual Diabetes Control Program 59 3.7 IoT 60 3.8 Expert Systems 62 3.9 Literature Review 63 3.10 Discussions 64 3.10.1 Process of developing type-2 and type 1 diabetes 66 3.10.2 Advantages of IoT use in individual diabetic control 66 3.10.3 Disadvantages of IoT use in individual diabetic control 68 3.11 Conclusion 74 3.12 Future Scope 76 4 Harnessing Machine Intelligence and Big Data for Diabetes Management 81 4.1 Introduction 82 4.2 Machine Learning in Diabetic Care 83 4.2.1 Supervised learning methodologies in diabetic disease diagnosis 84 4.2.1.1 Logistic regression: A precision tool for binary classification 86 4.2.1.2 Unveiling precision with support vector machines 86 4.2.1.3 Carving paths with decision trees in diabetic classification 87 4.2.1.4 Pioneering precision with neural networks 88 4.2.2 Unveiling the unsupervised: insights through data elevation 90 4.2.2.1 Clustering: Carving pathways to personalized diabetic care 90 4.2.2.2 Anomaly detection: Illuminating unseen risks in diabetic care 92 4.2.2.3 Dimensionality reduction: Navigating complexity in diabetic insight 93 4.3 The Enigmatic Influence of Big Data in Diabetic Care 94 4.4 Navigating Challenges and Seizing Opportunities in Diabetic Care 95 4.5 Ethical and Privacy Frontiers in Diabetes Management 96 4.6 Unleashing Transformation: Unveiling the Potential of Big Data and Machine Learning 98 4.7 Converging Horizons: Illuminating Emerging Trends and Technologies 99 4.8 Navigating the Uncharted: Forging Ahead in Diabetes Care 100 4.9 Conclusion 101 5 Machine Intelligence and Big Data in Diabetic Care: Laboratorian⁰́₉s Perspective 107 5.1 Introduction 108 5.2 Machine Intelligence in the Field of Laboratory Science 108 5.2.1 The significance of data in diabetic care 111 5.2.2 The role of big data analytics in diabetic care 111 5.2.3 The role of machine intelligence in diabetic care 111 5.2.4 The role of laboratory medicine in diabetic care 112 5.2.5 Personalized diabetes management 112 5.2.6 Early detection and prevention of diabetes complications 113 5.2.7 Predictive analytics for diabetes management 113 5.2.7.1 Results and impact 114 5.2.7.2 Challenges and considerations 115 5.2.8 Personalized treatment plans 116 5.2.9 Remote monitoring and telemedicine 119 5.2.10 Challenges and opportunities 119 5.3 Conclusion 120 6 EfficientNetB3-DTL: Classification of Diabetic Retinopathy Images using Modified EfficientNetB3 with Deep Transfer Learning 125 6.1 Introduction 126 6.2 Related Work 127 6.3 Proposed Model 129 6.3.1 Dataset 130 6.3.2 Data pre-processing 131 6.3.3 Deep transfer learning 132 6.4 Results and Discussion 137 6.5 Conclusion 141 7 Prediction and Diagnosis of Glaucoma in Fundus Images through Optic Cup and Optic Disk Segmentation 145 7.1 Introduction 146 7.2 Related Work 148 7.2.1 Color-based segmentation 149 7.2.2 Model-based segmentation 149 7.2.3 Texture-based segmentation 149 7.3 Material and Methods 151 7.3.1 Retinal image acquisition 151 7.3.2 Proposed methodology 151 7.3.3 Active contour segmentation for optic disc localization 152 7.3.4 Level set algorithm for optic cup localization 152 7.3.5 K-means clustering based segmentation algorithm 153 7.4 Feature Extraction on the Segmented Optic Disk and Cup 155 7.4.1 Cup to disc ratio (CDR) 155 7.4.2 Texture-based feature extraction 158 7.5 Experimental Analysis 159 7.5.1 Multi-class support vector machine (SVM) classifier 159 7.5.2 Extreme learning machine (ELM) classifier 161 7.5.3 Discussion 163 7.6 Conclusions and Future Work 164 8 Early Diagnosis of Diabetes using an Intelligent Machine Learning Technique 169 8.1 Introduction 170 8.2 Related Works 172 8.3 System Model and Problem Statement 174 8.4 Proposed Model 175 8.4.1 Process of the proposed methodology 176 8.4.1.1 Data training and preprocessing 176 8.4.1.2 Feature analysis 177 8.4.1.3 Classification and prediction 178 8.4.1.4 Gene expression analysis 178 8.5 Result and Discussion 181 8.5.1 Case study 181 8.5.2 Performance assessment 182 8.5.2.1 Precision 184 8.5.2.2 Accuracy 184 8.5.2.3 Recall 185 8.5.2.4 F-score 186 8.5.2.5 Error rate 186 8.5.2.6 Time 188 8.5.3 Discussion 188 8.6 Conclusion 189 9 Advanced Diabetes Prediction: A Comprehensive Analysis of Machine Learning and Deep Learning Techniques 195 9.1 Introduction 196 9.2 Related Work 202 9.2.1 Literature review on diabetes prediction using machine learning 203 9.2.2 Literature review on diabetes prediction using deep learning 204 9.3 Discussion 206 9.3.1 Datasets 206 9.3.2 Diabetes prediction using machine learning/deep learning techniques 206 9.4 Case Study 213 9.4.1 Collecting data 213 9.4.2 Preprocessing 213 9.4.3 Execution and outcomes 213 9.4.4 Analysis and examination 214 9.5 Summary 216 10 Intelligent Diagnosis Support System for Screening Diabetes Subjects using Hybrid Machine Learning Algorithms 223 10.1 Introduction 224 10.2 Related Work 225 10.3 Materials and Methods 228 10.3.1 Dataset 228 10.3.2 Data pre-processing 228 10.3.3 Machine learning models 230 10.4 Results and Discussion 233 10.5 Conclusion 237 11 Cyber⁰́₃Physical System for Managing Diabetic Healthcare 241 11.1 Background 242 11.1.1 Epidemiology of DM in India 242 11.1.2 Epidemiology of DM in USA 243 11.1.3 Limitations and future directions 245 11.1.4 Methods to prevent or reduce the effect of DM 245 11.1.5 Challenges of caregivers of patients suffering from DM 246 11.2 Materials 248 11.3 Methodology 250 11.4 Work Plan 252 11.4.1 Outcomes of diabetic healthcare system 258 11.5 Conclusion 259 Index 261 About the Editors 263.
