Advanced Healthcare Systems

ADVANCED HEALTHCARE SYSTEMS

THIS BOOK OFFERS A COMPLETE PACKAGE INVOLVING THE INCUBATION OF MACHINE LEARNING, AI, AND IOT IN HEALTHCARE THAT IS BENEFICIAL FOR RESEARCHERS, HEALTHCARE PROFESSIONALS, SCIENTISTS, AND TECHNOLOGISTS.The applications and challenges of machine learning and artificial intelligence in the Internet of Things (IoT) for healthcare applications are comprehensively covered in this book. IoT generates big data of varying data quality; intelligent processing and analysis of this big data are the keys to developing smart IoT applications, thereby making space for machine learning (ML) applications. Due to its computational tools that can substitute for human intelligence in the performance of certain tasks, artificial intelligence (AI) makes it possible for machines to learn from experience, adjust to new inputs and perform human-like tasks. Since IoT platforms provide an interface to gather data from various devices, they can easily be deployed into AI/ML systems. The value of AI in this context is its ability to quickly mesh insights from data and automatically identify patterns and detect anomalies in the data that smart sensors and devices generate—information such as temperature, pressure, humidity, air quality, vibration, and sound—that can be really helpful to rapid diagnosis. AUDIENCEThis book will be of interest to researchers in artificial intelligence, the Internet of Things, machine learning as well as information technologists working in the healthcare sector. ROHIT TANWAR, PHD (Kurukshetra University, Kurukshetra, India) is an assistant professor in the School of Computer Science at UPES Dehradun, India.

S. BALAMURUGAN, PHD, SMIEEE, ACM Distinguished Speaker, received his PhD from Anna University, India. He has published 57 books, 300+ international journals/conferences, and 100 patents. He is the Director of the Albert Einstein Engineering and Research Labs. He is also the Vice-Chairman of the Renewable Energy Society of India (RESI). He is serving as a research consultant to many companies, startups, SMEs, and MSMEs. He has received numerous awards for research at national and international levels. R. K. SAINI, PHD (DIT University, Dehradun, India) is an assistant professor in the Department of Computer Science & Applications at DIT University, Dehradun (Uttarakhand). VISHAL BHARTI, PHD is a professor in the Department of Computer Science and Engineering, Chandigarh University, India. He has published more than 75 research papers in both national & international journals. PREMKUMAR CHITHALURU, PHD is an assistant professor in the Department of SCS at the University of Petroleum and Energy Studies (UPES), Dehradun, India. Preface xvii

1 INTERNET OF MEDICAL THINGS—STATE-OF-THE-ART 1

Kishor Joshi and Ruchi Mehrotra

1.1 Introduction 2

1.2 Historical Evolution of IoT to IoMT 2

1.2.1 IoT and IoMT—Market Size 4

1.3 Smart Wearable Technology 4

1.3.1 Consumer Fitness Smart Wearables 4

1.3.2 Clinical-Grade Wearables 5

1.4 Smart Pills 7

1.5 Reduction of Hospital-Acquired Infections 8

1.5.1 Navigation Apps for Hospitals 8

1.6 In-Home Segment 8

1.7 Community Segment 9

1.8 Telehealth and Remote Patient Monitoring 9

1.9 IoMT in Healthcare Logistics and Asset Management 12

1.10 IoMT Use in Monitoring During COVID-19 13

1.11 Conclusion 14

References 15

2 ISSUES AND CHALLENGES RELATED TO PRIVACY AND SECURITY IN HEALTHCARE USING IOT, FOG, AND CLOUD COMPUTING 21

Hritu Raj, Mohit Kumar, Prashant Kumar, Amritpal Singh and Om Prakash Verma

2.1 Introduction 22

2.2 Related Works 23

2.3 Architecture 25

2.3.1 Device Layer 25

2.3.2 Fog Layer 26

2.3.3 Cloud Layer 26

2.4 Issues and Challenges 26

2.5 Conclusion 29

References 30

3 STUDY OF THYROID DISEASE USING MACHINE LEARNING 33

Shanu Verma, Rashmi Popli and Harish Kumar

3.1 Introduction 34

3.2 Related Works 34

3.3 Thyroid Functioning 35

3.4 Category of Thyroid Cancer 36

3.5 Machine Learning Approach Toward the Detection of Thyroid Cancer 37

3.5.1 Decision Tree Algorithm 38

3.5.2 Support Vector Machines 39

3.5.3 Random Forest 39

3.5.4 Logistic Regression 39

3.5.5 Naïve Bayes 40

3.6 Conclusion 41

References 41

4 A REVIEW OF VARIOUS SECURITY AND PRIVACY INNOVATIONS FOR IOT APPLICATIONS IN HEALTHCARE 43

Abhishek Raghuvanshi, Umesh Kumar Singh and Chirag Joshi

4.1 Introduction 44

4.1.1 Introduction to IoT 44

4.1.2 Introduction to Vulnerability, Attack, and Threat 45

4.2 IoT in Healthcare 46

4.2.1 Confidentiality 46

4.2.2 Integrity 46

4.2.3 Authorization 46

4.2.4 Availability 47

4.3 Review of Security and Privacy Innovations for IoT Applications in Healthcare, Smart Cities, and Smart Homes 48

4.4 Conclusion 54

References 54

5 METHODS OF LUNG SEGMENTATION BASED ON CT IMAGES 59

Amit Verma and Thipendra P. Singh

5.1 Introduction 59

5.2 Semi-Automated Algorithm for Lung Segmentation 60

5.2.1 Algorithm for Tracking to Lung Edge 60

5.2.2 Outlining the Region of Interest in CT Images 62

5.2.2.1 Locating the Region of Interest 62

5.2.2.2 Seed Pixels and Searching Outline 62

5.3 Automated Method for Lung Segmentation 63

5.3.1 Knowledge-Based Automatic Model for Segmentation 63

5.3.2 Automatic Method for Segmenting the Lung CT Image 64

5.4 Advantages of Automatic Lung Segmentation Over Manual and Semi-Automatic Methods 64

5.5 Conclusion 65

References 65

6 HANDLING UNBALANCED DATA IN CLINICAL IMAGES 69

Amit Verma

6.1 Introduction 70

6.2 Handling Imbalance Data 71

6.2.1 Cluster-Based Under-Sampling Technique 72

6.2.2 Bootstrap Aggregation (Bagging) 75

6.3 Conclusion 76

References 76

7 IOT-BASED HEALTH MONITORING SYSTEM FOR SPEECH-IMPAIRED PEOPLE USING ASSISTIVE WEARABLE ACCELEROMETER 81

Ishita Banerjee and Madhumathy P.

7.1 Introduction 82

7.2 Literature Survey 84

7.3 Procedure 86

7.4 Results 93

7.5 Conclusion 97

References 97

8 SMART IOT DEVICES FOR THE ELDERLY AND PEOPLE WITH DISABILITIES 101

K. N. D. Saile and Kolisetti Navatha

8.1 Introduction 101

8.2 Need for IoT Devices 102

8.3 Where Are the IoT Devices Used? 103

8.3.1 Home Automation 103

8.3.2 Smart Appliances 104

8.3.3 Healthcare 104

8.4 Devices in Home Automation 104

8.4.1 Automatic Lights Control 104

8.4.2 Automated Home Safety and Security 104

8.5 Smart Appliances 105

8.5.1 Smart Oven 105

8.5.2 Smart Assistant 105

8.5.3 Smart Washers and Dryers 106

8.5.4 Smart Coffee Machines 106

8.5.5 Smart Refrigerator 106

8.6 Healthcare 106

8.6.1 Smart Watches 107

8.6.2 Smart Thermometer 107

8.6.3 Smart Blood Pressure Monitor 107

8.6.4 Smart Glucose Monitors 107

8.6.5 Smart Insulin Pump 108

8.6.6 Smart Wearable Asthma Monitor 108

8.6.7 Assisted Vision Smart Glasses 109

8.6.8 Finger Reader 109

8.6.9 Braille Smart Watch 109

8.6.10 Smart Wand 109

8.6.11 Taptilo Braille Device 110

8.6.12 Smart Hearing Aid 110

8.6.13 E-Alarm 110

8.6.14 Spoon Feeding Robot 110

8.6.15 Automated Wheel Chair 110

8.7 Conclusion 112

References 112

9 IOT-BASED HEALTH MONITORING AND TRACKING SYSTEM FOR SOLDIERS 115

Kavitha N. and Madhumathy P.

9.1 Introduction 116

9.2 Literature Survey 117

9.3 System Requirements 118

9.3.1 Software Requirement Specification 119

9.3.2 Functional Requirements 119

9.4 System Design 119

9.4.1 Features 121

9.4.1.1 On-Chip Flash Memory 122

9.4.1.2 On-Chip Static RAM 122

9.4.2 Pin Control Block 122

9.4.3 UARTs 123

9.4.3.1 Features 123

9.4.4 System Control 123

9.4.4.1 Crystal Oscillator 123

9.4.4.2 Phase-Locked Loop 124

9.4.4.3 Reset and Wake-Up Timer 124

9.4.4.4 Brown Out Detector 125

9.4.4.5 Code Security 125

9.4.4.6 External Interrupt Inputs 125

9.4.4.7 Memory Mapping Control 125

9.4.4.8 Power Control 126

9.4.5 Real Monitor 126

9.4.5.1 GPS Module 126

9.4.6 Temperature Sensor 127

9.4.7 Power Supply 128

9.4.8 Regulator 128

9.4.9 LCD 128

9.4.10 Heart Rate Sensor 129

9.5 Implementation 129

9.5.1 Algorithm 130

9.5.2 Hardware Implementation 130

9.5.3 Software Implementation 131

9.6 Results and Discussions 133

9.6.1 Heart Rate 133

9.6.2 Temperature Sensor 135

9.6.3 Panic Button 135

9.6.4 GPS Receiver 135

9.7 Conclusion 136

References 136

10 CLOUD-IOT SECURED PREDICTION SYSTEM FOR PROCESSING AND ANALYSIS OF HEALTHCARE DATA USING MACHINE LEARNING TECHNIQUES 137

G. K. Kamalam and S. Anitha

10.1 Introduction 138

10.2 Literature Survey 139

10.3 Medical Data Classification 141

10.3.1 Structured Data 142

10.3.2 Semi-Structured Data 142

10.4 Data Analysis 142

10.4.1 Descriptive Analysis 142

10.4.2 Diagnostic Analysis 143

10.4.3 Predictive Analysis 143

10.4.4 Prescriptive Analysis 143

10.5 ML Methods Used in Healthcare 144

10.5.1 Supervised Learning Technique 144

10.5.2 Unsupervised Learning 145

10.5.3 Semi-Supervised Learning 145

10.5.4 Reinforcement Learning 145

10.6 Probability Distributions 145

10.6.1 Discrete Probability Distributions 146

10.6.1.1 Bernoulli Distribution 146

10.6.1.2 Uniform Distribution 147

10.6.1.3 Binomial Distribution 147

10.6.1.4 Normal Distribution 148

10.6.1.5 Poisson Distribution 148

10.6.1.6 Exponential Distribution 149

10.7 Evaluation Metrics 150

10.7.1 Classification Accuracy 150

10.7.2 Confusion Matrix 150

10.7.3 Logarithmic Loss 151

10.7.4 Receiver Operating Characteristic Curve, or ROC Curve 152

10.7.5 Area Under Curve (AUC) 152

10.7.6 Precision 153

10.7.7 Recall 153

10.7.8 F1 Score 153

10.7.9 Mean Absolute Error 154

10.7.10 Mean Squared Error 154

10.7.11 Root Mean Squared Error 155

10.7.12 Root Mean Squared Logarithmic Error 155

10.7.13 R-Squared/Adjusted R-Squared 156

10.7.14 Adjusted R-Squared 156

10.8 Proposed Methodology 156

10.8.1 Neural Network 158

10.8.2 Triangular Membership Function 158

10.8.3 Data Collection 159

10.8.4 Secured Data Storage 159

10.8.5 Data Retrieval and Merging 161

10.8.6 Data Aggregation 162

10.8.7 Data Partition 162

10.8.8 Fuzzy Rules for Prediction of Heart Disease 163

10.8.9 Fuzzy Rules for Prediction of Diabetes 164

10.8.10 Disease Prediction With Severity and Diagnosis 165

10.9 Experimental Results 166

10.10 Conclusion 169

References 169

11 CLOUDIOT-DRIVEN HEALTHCARE: REVIEW, ARCHITECTURE, SECURITY IMPLICATIONS, AND OPEN RESEARCH ISSUES 173

Junaid Latief Shah, Heena Farooq Bhat and Asif Iqbal Khan

11.1 Introduction 174

11.2 Background Elements 180

11.2.1 Security Comparison Between Traditional and IoT Networks 185

11.3 Secure Protocols and Enabling Technologies for CloudIoT Healthcare Applications 187

11.3.1 Security Protocols 187

11.3.2 Enabling Technologies 188

11.4 CloudIoT Health System Framework 191

11.4.1 Data Perception/Acquisition 192

11.4.2 Data Transmission/Communication 193

11.4.3 Cloud Storage and Warehouse 194

11.4.4 Data Flow in Healthcare Architecture - A Conceptual Framework 194

11.4.5 Design Considerations 197

11.5 Security Challenges and Vulnerabilities 199

11.5.1 Security Characteristics and Objectives 200

11.5.1.1 Confidentiality 202

11.5.1.2 Integrity 202

11.5.1.3 Availability 202

11.5.1.4 Identification and Authentication 202

11.5.1.5 Privacy 203

11.5.1.6 Light Weight Solutions 203

11.5.1.7 Heterogeneity 203

11.5.1.8 Policies 203

11.5.2 Security Vulnerabilities 203

11.5.2.1 IoT Threats and Vulnerabilities 205

11.5.2.2 Cloud-Based Threats 208

11.6 Security Countermeasures and Considerations 214

11.6.1 Security Countermeasures 214

11.6.1.1 Security Awareness and Survey 214

11.6.1.2 Security Architecture and Framework 215

11.6.1.3 Key Management 216

11.6.1.4 Authentication 217

11.6.1.5 Trust 218

11.6.1.6 Cryptography 219

11.6.1.7 Device Security 219

11.6.1.8 Identity Management 220

11.6.1.9 Risk-Based Security/Risk Assessment 220

11.6.1.10 Block Chain–Based Security 220

11.6.1.11 Automata-Based Security 220

11.6.2 Security Considerations 234

11.7 Open Research Issues and Security Challenges 237

11.7.1 Security Architecture 237

11.7.2 Resource Constraints 238

11.7.3 Heterogeneous Data and Devices 238

11.7.4 Protocol Interoperability 238

11.7.5 Trust Management and Governance 239

11.7.6 Fault Tolerance 239

11.7.7 Next-Generation 5G Protocol 240

11.8 Discussion and Analysis 240

11.9 Conclusion 241

References 242

12 A NOVEL USAGE OF ARTIFICIAL INTELLIGENCE AND INTERNET OF THINGS IN REMOTE-BASED HEALTHCARE APPLICATIONS 255

V. Arulkumar, D. Mansoor Hussain, S. Sridhar and P. Vivekanandan

12.1 Introduction Machine Learning 256

12.2 Importance of Machine Learning 256

12.2.1 ML vs. Classical Algorithms 258

12.2.2 Learning Supervised 259

12.2.3 Unsupervised Learning 261

12.2.4 Network for Neuralism 263

12.2.4.1 Definition of the Neural Network 263

12.2.4.2 Neural Network Elements 263

12.3 Procedure 265

12.3.1 Dataset and Seizure Identification 265

12.3.2 System 265

12.4 Feature Extraction 266

12.5 Experimental Methods 266

12.5.1 Stepwise Feature Optimization 266

12.5.2 Post-Classification Validation 268

12.5.3 Fusion of Classification Methods 268

12.6 Experiments 269

12.7 Framework for EEG Signal Classification 269

12.8 Detection of the Preictal State 270

12.9 Determination of the Seizure Prediction Horizon 271

12.10 Dynamic Classification Over Time 272

12.11 Conclusion 273

References 273

13 USE OF MACHINE LEARNING IN HEALTHCARE 275

V. Lakshman Narayana, R. S. M. Lakshmi Patibandla, B. Tarakeswara Rao and Arepalli Peda Gopi

13.1 Introduction 276

13.2 Uses of Machine Learning in Pharma and Medicine 276

13.2.1 Distinguish Illnesses and Examination 277

13.2.2 Drug Discovery and Manufacturing 277

13.2.3 Scientific Imaging Analysis 278

13.2.4 Twisted Therapy 278

13.2.5 AI to Know-Based Social Change 278

13.2.6 Perception Wellness Realisms 279

13.2.7 Logical Preliminary and Exploration 279

13.2.8 Publicly Supported Perceptions Collection 279

13.2.9 Better Radiotherapy 280

13.2.10 Incidence Forecast 280

13.3 The Ongoing Preferences of ML in Human Services 281

13.4 The Morals of the Use of Calculations in Medicinal Services 284

13.5 Opportunities in Healthcare Quality Improvement 288

13.5.1 Variation in Care 288

13.5.2 Inappropriate Care 289

13.5.3 Prevents Care–Associated Injurious and Death for Carefrontation 289

13.5.4 The Fact That People Are Unable to do What They Know Works 289

13.5.5 A Waste 290

13.6 A Team-Based Care Approach Reduces Waste 290

13.7 Conclusion 291

References 292

14 METHODS OF MRI BRAIN TUMOR SEGMENTATION 295

Amit Verma

14.1 Introduction 295

14.2 Generative and Descriptive Models 296

14.2.1 Region-Based Segmentation 300

14.2.2 Generative Model With Weighted Aggregation 300

14.3 Conclusion 302

References 303

15 EARLY DETECTION OF TYPE 2 DIABETES MELLITUS USING DEEP NEURAL NETWORK–BASED MODEL 305

Varun Sapra and Luxmi Sapra

15.1 Introduction 306

15.2 Data Set 307

15.2.1 Data Insights 308

15.3 Feature Engineering 310

15.4 Framework for Early Detection of Disease 312

15.4.1 Deep Neural Network 313

15.5 Result 314

15.6 Conclusion 315

References 315

16 A COMPREHENSIVE ANALYSIS ON MASKED FACE DETECTION ALGORITHMS 319

Pranjali Singh, Amitesh Garg and Amritpal Singh

16.1 Introduction 320

16.2 Literature Review 321

16.3 Implementation Approach 325

16.3.1 Feature Extraction 325

16.3.2 Image Processing 325

16.3.3 Image Acquisition 325

16.3.4 Classification 325

16.3.5 MobileNetV2 326

16.3.6 Deep Learning Architecture 326

16.3.7 LeNet-5, AlexNet, and ResNet-50 326

16.3.8 Data Collection 326

16.3.9 Development of Model 327

16.3.10 Training of Model 328

16.3.11 Model Testing 328

16.4 Observation and Analysis 328

16.4.1 CNN Algorithm 328

16.4.2 SSDNETV2 Algorithm 330

16.4.3 SVM 331

16.5 Conclusion 332

References 333

17 IOT-BASED AUTOMATED HEALTHCARE SYSTEM 335

Darpan Anand and Aashish Kumar

17.1 Introduction 335

17.1.1 Software-Defined Network 336

17.1.2 Network Function Virtualization 337

17.1.3 Sensor Used in IoT Devices 338

17.2 SDN-Based IoT Framework 341

17.3 Literature Survey 343

17.4 Architecture of SDN-IoT for Healthcare System 344

17.5 Challenges 345

17.6 Conclusion 347

References 347

Index 351