Medical Imaging and Health Informatics
Description
MEDICAL IMAGING AND HEALTH INFORMATICS
Provides a comprehensive review of artificial intelligence (AI) in medical imaging as well as practical recommendations for the usage of machine learning (ML) and deep learning (DL) techniques for clinical applications.
Medical imaging and health informatics is a subfield of science and engineering which applies informatics to medicine and includes the study of design, development, and application of computational innovations to improve healthcare. The health domain has a wide range of challenges that can be addressed using computational approaches; therefore, the use of AI and associated technologies is becoming more common in society and healthcare. Currently, deep learning algorithms are a promising option for automated disease detection with high accuracy. Clinical data analysis employing these deep learning algorithms allows physicians to detect diseases earlier and treat patients more efficiently. Since these technologies have the potential to transform many aspects of patient care, disease detection, disease progression and pharmaceutical organization, approaches such as deep learning algorithms, convolutional neural networks, and image processing techniques are explored in this book.
This book also delves into a wide range of image segmentation, classification, registration, computer-aided analysis applications, methodologies, algorithms, platforms, and tools; and gives a holistic approach to the application of AI in healthcare through case studies and innovative applications. It also shows how image processing, machine learning and deep learning techniques can be applied for medical diagnostics in several specific health scenarios such as COVID-19, lung cancer, cardiovascular diseases, breast cancer, liver tumor, bone fractures, etc. Also highlighted are the significant issues and concerns regarding the use of AI in healthcare together with other allied areas, such as the Internet of Things (IoT) and medical informatics, to construct a global multidisciplinary forum.
Audience The core audience comprises researchers and industry engineers, scientists, radiologists, healthcare professionals, data scientists who work in health informatics, computer vision and medical image analysis.
More details
Other editions
Additional editions
Persons
Tushar H. Jaware, PhD, received his degree in Medical Image Processing and is now an assistant professor in the Department of Electronics and Telecommunication Engineering, R C Patel Institute of Technology, Shirpur, India. He has published more than 50 research articles in refereed journals and IEEE conferences, and has three international patents granted and two Indian patents published.
K. Sarat Kumar, PhD, received his degree in Electronics Engineering and is now a professor in the Department of Electronics & Communication Engineering, K L University, Andhra Pradesh, India.
Ravindra D. Badgujar, PhD, received his degree in Electronics Engineering and is now an assistant professor in the Department of Electronics and Telecommunication Engineering, R C Patel Institute of Technology, Shirpur, India. He has published many research articles in refereed journals and IEEE conferences as well as one international patent granted and two Indian patents published.
Svetlin Antonov, PhD, received his degree in Telecommunications and is now a lecturer in the Faculty of Telecommunications, TU-Sofia, Bulgaria. He is the author of several books and more than 60 peer-reviewed articles.
Content
Preface xvii
1 Machine Learning Approach for Medical Diagnosis Based on Prediction Model 1
Hemant Kasturiwale, Rajesh Karhe and Sujata N. Kale
1.1 Introduction 2
1.1.1 Heart System and Major Cardiac Diseases 2
1.1.2 ECG for Heart Rate Variability Analysis 2
1.1.3 HRV for Cardiac Analysis 3
1.2 Machine Learning Approach and Prediction 3
1.3 Material and Experimentation 4
1.3.1 Data and HRV 4
1.3.1.1 HRV Data Analysis via ECG Data Acquisition System 5
1.3.2 Methodology and Techniques 6
1.3.2.1 Classifiers and Performance Evaluation 7
1.3.3 Proposed Model With Layer Representation 8
1.3.4 The Model Using Fixed Set of Features and Standard Dataset 11
1.3.4.1 Performance of Classifiers With Feature Selection 11
1.4 Performance Metrics and Evaluation of Classifiers 13
1.4.1 Cardiac Disease Prediction Through Flexi Intra Group Selection Model 13
1.4.2 HRV Model With Flexi Set of Features 14
1.4.3 Performance of the Proposed Modified With ISM-24 15
1.5 Discussion and Conclusion 18
1.5.1 Conclusion and Future Scope 19
References 20
2 Applications of Machine Learning Techniques in Disease Detection 23
M.S. Roobini, Sowmiya M., S. Jancy and L. Suji Helen
2.1 Introduction 24
2.1.1 Overview of Machine Learning Types 24
2.1.2 Motivation 25
2.1.3 Organization the Chapter 25
2.2 Types of Machine Learning Techniques 25
2.2.1 Supervised Learning 25
2.2.2 Classification Algorithm 25
2.2.3 Regression Analysis 26
2.2.4 Linear Regression 27
2.2.4.1 Applications of Linear Regression 27
2.2.5 KNN Algorithm 28
2.2.5.1 Working of KNN 28
2.2.5.2 Drawbacks of KNN Algorithm 29
2.2.6 Decision Tree Classification Algorithm 29
2.2.6.1 Attribute Selection Measures 29
2.2.6.2 Information Gain 29
2.2.6.3 Gain Ratio 29
2.2.7 Random Forest Algorithm 29
2.2.7.1 How the Random Forest Algorithm Works 29
2.2.7.2 Advantage of Using Random Forest 30
2.2.7.3 Disadvantage of Using the Random Forest 31
2.2.8 Naive Bayes Classifier Algorithm 31
2.2.8.1 For What Reason is it Called Naive Bayes? 31
2.2.8.2 Disservices of Naive Bayes Classifier 31
2.2.9 Logistic Regression 31
2.2.9.1 Logistic Regression for Machine Learning 31
2.2.10 Support Vector Machine 32
2.2.11 Unsupervised Learning 32
2.2.11.1 Clustering 33
2.2.11.2 PCA in Machine Learning 35
2.2.12 Semi-Supervised Learning 38
2.2.12.1 What is Semi-Supervised Clustering? 38
2.2.12.2 How Semi-Supervised Learning Functions? 38
2.2.13 Reinforcement Learning 39
2.2.13.1 Artificial Intelligence 39
2.2.13.2 Deep Learning 40
2.2.13.3 Points of Interest of Machine Learning 41
2.2.13.4 Why Machine Learning is Popular 41
2.2.13.5 Test Utilizations of ML 42
2.3 Future Research Directions 43
2.3.1 Privacy 43
2.3.2 Accuracy 43
References 43
3 Dengue Incidence Rate Prediction Using Nonlinear Autoregressive Neural Network Time Series Model 47
S. Dhamodharavadhani and R. Rathipriya
3.1 Introduction 47
3.2 Related Literature Study 48
3.2.1 Limitations of Existing Works 50
3.2.2 Contributions of Proposed Methodology 50
3.3 Methods and Materials 50
3.3.1 NAR-NNTS 50
3.3.2 Fit/Train the Model 51
3.3.3 Training Algorithms 54
3.3.3.1 Levenberg-Marquardt (LM) Algorithm 54
3.3.3.2 Bayesian Regularization (BR) Algorithm 55
3.3.3.3 Scaled Conjugate Gradient (SCG) Algorithm 55
3.3.4 DIR Prediction 55
3.4 Result Discussions 56
3.4.1 Dataset Description 56
3.4.2 Evaluation Measure for NAR-NNTS Models 57
3.4.3 Analysis of Results 57
3.5 Conclusion and Future Work 65
Acknowledgment 66
References 66
4 Early Detection of Breast Cancer Using Machine Learning 69
G. Lavanya and G. Thilagavathi
4.1 Introduction 70
4.1.1 Objective 70
4.1.2 Anatomy of Breast 70
4.1.3 Breast Imaging Modalities 71
4.2 Methodology 71
4.2.1 Database 71
4.2.2 Image Pre-Processing 71
4.3 Segmentation 72
4.4 Feature Extraction 72
4.5 Classification 72
4.5.1 Naive Bayes Neural Network Classifier 72
4.5.2 Radial Basis Function Neural Network 73
4.5.2.1 Input 73
4.5.2.2 Hidden Layer 73
4.5.2.3 Output Nodes 74
4.6 Performance Evaluation Methods 74
4.7 Output 75
4.7.1 Dataset 75
4.7.2 Pre-Processing 75
4.7.3 Segmentation 75
4.7.4 Geometric Feature Extraction 77
4.8 Results and Discussion 78
4.8.1 Database 78
4.9 Conclusion and Future Scope 81
References 81
5 Machine Learning Approach for Prediction of Lung Cancer 83
Hemant Kasturiwale, Swati Bhisikar and Sandhya Save
5.1 Introduction 84
5.1.1 Disorders in Lungs 84
5.1.2 Background 84
5.1.3 Material, Datasets, and Techniques 85
5.2 Feature Extraction and Lung Cancer Analysis 86
5.3 Methodology 87
5.3.1 Proposed Algorithm Steps 87
5.3.2 Classifiers in Concurrence With Datasets 88
5.4 Proposed System and Implementation 89
5.4.1 Interpretation via Artificial Intelligence 89
5.4.2 Training of Model 90
5.4.3 Implementation and Results 90
5.5 Conclusion 99
5.5.1 Future Scope 99
References 100
6 Segmentation of Liver Tumor Using ANN 103
Hema L. K. and R. Indumathi
6.1 Introduction 103
6.2 Liver Tumor 104
6.2.1 Overview of Liver Tumor 104
6.2.2 Classification 105
6.2.2.1 Benign 105
6.2.2.2 Malignant 107
6.3 Benefits of CT to Diagnose Liver Cancer 108
6.4 Literature Review 108
6.5 Interactive Liver Tumor Segmentation by Deep Learning 109
6.6 Existing System 109
6.7 Proposed System 110
6.7.1 Pre-Processing 110
6.7.2 Segmentation 111
6.7.3 Feature Extraction 112
6.7.4 GLCM 112
6.7.5 Backpropagation Network 113
6.8 Result and Discussion 113
6.8.1 Processed Images 114
6.8.2 Segmentation 116
6.9 Future Enhancements 117
6.10 Conclusion 118
References 118
7 DMSAN: Deep Multi-Scale Attention Network for Automatic Liver Segmentation From Abdomen CT Images 121
Devidas T. Kushnure and Sanjay N. Talbar
7.1 Introduction 121
7.2 Related Work 122
7.3 Methodology 123
7.3.1 Proposed Architecture 123
7.3.2 Multi-Scale Feature Characterization Using Res2Net Module 125
7.4 Experimental Analysis 126
7.4.1 Dataset Description 126
7.4.2 Pre-Processing Dataset 127
7.4.3 Training Strategy 128
7.4.4 Loss Function 128
7.4.5 Implementation Platform 129
7.4.6 Data Augmentation 129
7.4.7 Performance Metrics 129
7.5 Results 131
7.6 Result Comparison With Other Methods 135
7.7 Discussion 136
7.8 Conclusion 137
Acknowledgement 138
References 138
8 AI-Based Identification and Prediction of Cardiac Disorders 141
Rajesh Karhe, Hemant Kasturiwale and Sujata N. Kale
8.1 Introduction 142
8.1.1 Cardiac Electrophysiology and Electrocardiogram 143
8.1.2 Heart Arrhythmia 144
8.1.2.1 Types of Arrhythmias 145
8.1.3 ECG Database 147
8.1.3.1 Association for the Advancement of Medical Instrumentation (AAMI) Standard 147
8.1.4 An Overview of ECG Signal Analysis 148
8.2 Related Work 149
8.3 Classifiers and Methodology 151
8.3.1 Databases for Cardiac Arrhythmia Detection 152
8.3.2 MIT-BIH Normal Sinus Rhythm and Arrhythmia Database 152
8.3.3 Arrhythmia Detection and Classification 153
8.3.4 Methodology 153
8.3.4.1 Database Gathering and Pre-Processing 153
8.3.4.2 QRST Wave Detection 153
8.3.4.3 Features Extraction 154
8.3.4.4 Neural Network 155
8.3.4.5 Performance Evaluation 156
8.4 Result Analysis 156
8.4.1 Arrhythmia Detection and Classification 156
8.4.2 Dataset 156
8.4.3 Evaluations and Results 156
8.4.4 Evaluating the Performance of Various Neural Network Classifiers (Arrhythmia Detection) 157
8.5 Conclusions and Future Scope 159
8.5.1 Arrhythmia Detection and Classification 159
8.5.2 Future Scope 161
References 161
9 An Implementation of Image Processing Technique for Bone Fracture Detection Including Classification 165
Rocky Upadhyay, Prakash Singh Tanwar and Sheshang Degadwala
9.1 Introduction 165
9.2 Existing Technology 166
9.2.1 Pre-Processing 166
9.2.2 Denoise Image 167
9.2.3 Histogram 168
9.3 Image Processing 169
9.3.1 Canny Edge 169
9.4 Overview of System and Steps 170
9.4.1 Workflow 170
9.4.2 Classifiers 171
9.4.2.1 Extra Tree Ensemble Method 171
9.4.2.2 SVM 172
9.4.2.3 Trained Algorithm 173
9.4.3 Feature Extraction 173
9.5 Results 174
9.5.1 Result Analysis 175
9.6 Conclusion 176
References 176
10 Improved Otsu Algorithm for Segmentation of Malaria Parasite Images 179
Mosam K. Sangole, Sanjay T. Gandhe and Dipak P. Patil
10.1 Introduction 179
10.2 Literature Review 180
10.3 Related Works 182
10.4 Proposed Algorithm 183
10.5 Experimental Results 184
10.6 Conclusion 193
References 193
11 A Reliable and Fully Automated Diagnosis of COVID-19 Based on Computed Tomography 195
Bramah Hazela, Saad Bin Khalid and Pallavi Asthana
11.1 Introduction 196
11.2 Background 196
11.3 Methodology 199
11.3.1 Models Used 199
11.3.2 Architecture of the Image Source Classification Model 199
11.3.3 Architecture of the CT Scan Classification Model 200
11.3.4 Architecture of the Ultrasound Image Classification Model 201
11.3.5 Architecture of the X-Ray Classification Model 201
11.3.6 Dataset 202
11.3.6.1 Training 202
11.4 Results 204
11.5 Conclusion 206
References 207
12 Multimodality Medical Images for Healthcare Disease Analysis 209
B. Rajalingam, R. Santhoshkumar, P. Santosh Kumar Patra, M. Narayanan, G. Govinda Rajulu and T. Poongothai
12.1 Introduction 210
12.1.1 Background 210
12.2 Brief Survey of Earlier Works 212
12.3 Medical Imaging Modalities 213
12.3.1 Computed Tomography (CT) 214
12.3.2 Magnetic Resonance Imaging (MRI) 214
12.3.3 Positron Emission Tomography (PET) 214
12.3.4 Single-Photon Emission Computed Tomography (SPECT) 215
12.4 Image Fusion 216
12.4.1 Different Levels of Image Fusion 216
12.4.1.1 Pixel Level Fusion 216
12.4.1.2 Feature Level Fusion 217
12.4.1.3 Decision Level Fusion 217
12.5 Clinical Relevance for Medical Image Fusion 218
12.5.1 Clinical Relevance for Neurocyticercosis (NCC) 218
12.5.2 Clinical Relevance for Neoplastic Disease 218
12.5.2.1 Clinical Relevance for Astrocytoma 218
12.5.2.2 Clinical Relevance for Anaplastic Astrocytoma 219
12.5.2.3 Clinical Relevance for Metastatic Bronchogenic Carcinoma 220
12.5.3 Clinical Relevance for Alzheimer's Disease 221
12.6 Data Sets and Softwares Used 221
12.7 Generalized Image Fusion Scheme 221
12.7.1 Input Image Modalities 222
12.7.2 Image Registration 222
12.7.3 Fusion Process 223
12.7.4 Fusion Rule 223
12.7.5 Evaluation 224
12.7.5.1 Subjective Evaluation 224
12.7.5.2 Objective Evaluation 224
12.8 Medical Image Fusion Methods 224
12.8.1 Traditional Image Fusion Techniques 224
12.8.1.1 Spatial Domain Image Fusion Approach 225
12.8.1.2 Transform Domain Image Fusion Approach 225
12.8.1.3 Fuzzy Logic-Based Image Fusion Approach 227
12.8.1.4 Filtering Technique-Based Image Fusion Approach 227
12.8.1.5 Neural Network-Based Image Fusion Approach 227
12.8.2 Hybrid Image Fusion Techniques 228
12.8.2.1 Transforms with Fuzzy Logic-Based Medical Image Fusion 228
12.8.2.2 Transforms With Guided Image Filtering-Based Medical Image Fusion 229
12.8.2.3 Transforms With Neural Network-Based Image Fusion 229
12.9 Conclusions 233
12.9.1 Future Work 234
References 234
13 Health Detection System for COVID-19 Patients Using IoT 237
Dipak P. Patil, Kishor Badane, Amit Kumar Mishra and Vishal A. Wankhede
13.1 Introduction 237
13.1.1 Overview 237
13.1.2 Preventions 238
13.1.3 Symptoms 238
13.1.4 Present Situation 238
13.2 Related Works 239
13.3 System Design 239
13.3.1 Hardware Implementation 239
13.3.1.1 NodeMCU 240
13.3.1.2 DHT 11 Sensor 240
13.3.1.3 MAX30100 Oxygen Sensor 241
13.3.1.4 ThingSpeak Server 242
13.3.1.5 Arduino IDE 243
13.4 Proposed System for Detection of Corona Patients 245
13.4.1 Introduction 245
13.4.2 Arduino IDE 246
13.4.3 Hardware Implementation 246
13.5 Results and Performance Analysis 247
13.5.1 Hardware Implementation 247
13.5.1.1 Implementation of NodeMCU With Temperature Sensor 247
13.5.2 Software Implementation 248
13.5.2.1 Simulation of Temperature Sensor With Arduino on Proteus Software 248
13.5.2.2 Interfacing of LCD With Arduino 250
13.6 Conclusion 250
References 250
14 Intelligent Systems in Healthcare 253
Rajiv Dey and Pankaj Sahu
14.1 Introduction 253
14.2 Brain Computer Interface 255
14.2.1 Types of Signals Used in BCI 256
14.2.2 Components of BCI 257
14.2.3 Applications of BCI in Health Monitoring 258
14.3 Robotic Systems 258
14.3.1 Advantages of Surgical Robots 258
14.3.2 Centralization of the Important Information to the Surgeon 259
14.3.3 Remote-Surgery, Software Development, and High Speed
Connectivity Such as 5G 260
14.4 Voice Recognition Systems 260
14.5 Remote Health Monitoring Systems 260
14.5.1 Tele-Medicine Health Concerns 262
14.6 Internet of Things-Based Intelligent Systems 262
14.6.1 Ubiquitous Computing Technologies in Healthcare 264
14.6.2 Patient Bio-Signals and Acquisition Methods 265
14.6.3 Communication Technologies Used in Healthcare Application 267
14.6.4 Communication Technologies Based on Location/Position 269
14.7 Intelligent Electronic Healthcare Systems 270
14.7.1 The Background of Electronic Healthcare Systems 270
14.7.2 Intelligent Agents in Electronic Healthcare System 270
14.7.3 Patient Data Classification Techniques 271
14.8 Conclusion 271
References 272
15 Design of Antennas for Microwave Imaging Techniques 275
Dnyaneshwar D. Ahire, Gajanan K. Kharate and Ammar Muthana
15.1 Introduction 275
15.1.1 Overview 276
15.2 Literature 277
15.2.1 Microstrip Patch Antenna 278
15.2.2 Early Detection of Breast Cancer and Microstrip Patch Antenna for Biomedical Application 279
15.2.3 UWB for Microwave Imaging 279
15.3 Design and Development of Wideband Antenna 280
15.3.1 Overview 280
15.3.2 Design of Rectangular Microstrip Patch Antenna 281
15.3.3 Design of Microstrip Line Feed Rectangular Microstrip Patch Antenna 283
15.3.4 Design of Microstrip Line Feed Rectangular Microstrip Patch Antenna With Partial Ground 285
15.3.5 Key Shape Monopole Rectangular Microstrip Patch Antenna With Rounded Corner in Partial Ground 286
15.4 Results and Inferences 290
15.4.1 Overview 290
15.4.2 Rectangular Microstrip Patch Antenna 290
15.4.2.1 Reflection and VSWR Bandwidth 290
15.4.2.2 Surface Current Distribution 291
15.4.3 Microstrip Line Feed Rectangular Microstrip Patch Antenna With Partial Ground 292
15.4.3.1 Reflection and VSWR Bandwidth 292
15.4.3.2 Surface Current Distribution 292
15.4.3.3 Inference 293
15.4.4 Key Shape Monopole Rectangular Microstrip Patch Antenna with Rounded Corner in Partial Ground 294
15.4.4.1 Reflection and VSWR Bandwidth 294
15.4.4.2 Surface Current Distribution 294
15.4.4.3 Results of the Fabricated Antenna 295
15.4.4.4 Inference 296
15.5 Conclusion 297
References 298
16 COVID-19: A Global Crisis 303
Savita Mandan and Durgeshwari Kalal
16.1 Introduction 303
16.1.1 Structure 304
16.1.2 Classification of Corona Virus 304
16.1.3 Types of Human Coronavirus 304
16.1.4 Genome Organization of Corona Virus 305
16.1.5 Coronavirus Replication 305
16.1.6 Host Defenses 306
16.2 Clinical Manifestation and Pathogenesis 306
16.2.1 Symptoms 307
16.2.2 Epidemiology 307
16.3 Diagnosis and Control 308
16.3.1 Molecular Test 308
16.3.2 Serology 308
16.3.3 Concerning Lab Assessments 309
16.3.4 Significantly Improved D-Dimer 309
16.3.5 Imaging 309
16.3.6 HRCT 309
16.3.7 Lung Ultrasound 310
16.4 Control Measures 310
16.4.1 Prevention and Patient Education 311
16.5 Immunization 312
16.5.1 Medications 312
16.6 Conclusion 313
References 313
17 Smart Healthcare for Pregnant Women in Rural Areas 317
D. Shanthi
17.1 Introduction 317
17.2 National/International Surveys Reviews 319
17.2.1 National Family Health Survey Review-11 319
17.2.2 National Family Health Survey Review-2.2 319
17.2.3 National Family Health Survey Reviews-3 320
17.3 Architecture 320
17.4 Anganwadi's Collaborative Work 321
17.5 Schemes Offered by Central/State Governments 321
17.5.1 AAH (Anna Amrutha Hastham) 321
17.5.2 Programme Arogya Laxmi 323
17.5.3 Balamrutham-Kids' Weaning Food from 7 Months to 3 Years 323
17.5.4 Nutri TASC (Tracking of Group Responsibility for Services) 323
17.5.5 Akshyapatra Foundation (ISKCON) 324
17.5.6 Mahila Sishu Chaitanyam 324
17.5.7 Community Management of Acute Malnutrition 325
17.5.8 Child Health Nutrition Committee 325
17.5.9 Bharat Ratna APJ Abdul Kalam Amrut Yojna 325
17.6 Smart Healthcare System 326
17.7 Data Collection 328
17.8 Hardware and Software Features of HCS 328
17.9 Implementation 329
17.9.1 Modules 329
17.9.2 Modules Description 329
17.9.2.1 Data Preprocessing 329
17.9.2.2 Component Features Extraction 329
17.9.2.3 User Sentimental Measurement 330
17.9.2.4 Sentiment Evaluation 330
17.10 Results and Analysis 331
17.11 Conclusion 333
References 333
18 Computer-Aided Interpretation of ECG Signal-A Challenge 335
Shalini Sahay and A.K. Wadhwani
18.1 Introduction 336
18.1.1 Electrical Activity of the Heart 336
18.2 The Cardiovascular System 338
18.3 Electrocardiogram Leads 340
18.4 Artifacts/Noises Affecting the ECG 342
18.4.1 Baseline Wander 343
18.4.2 Power Line Interference 343
18.4.3 Motion Artifacts 344
18.4.4 Muscle Noise 344
18.4.5 Instrumentation Noise 344
18.4.6 Other Interferences 345
18.5 The ECG Waveform 346
18.5.1 Normal Sinus Rhythm 347
18.6 Cardiac Arrhythmias 347
18.6.1 Sinus Bradycardia 347
18.6.2 Sinus Tachycardia 348
18.6.3 Atrial Flutter 348
18.6.4 Atrial Fibrillation 349
18.6.5 Ventric ular Tachycardia 349
18.6.6 AV Block 2 First Degree 350
18.6.7 Asystole 350
18.7 Electrocardiogram Databases 351
18.8 Computer-Aided Interpretation (CAD) 351
18.9 Computational Techniques 354
18.10 Conclusion 356
References 357
Index 359
1
Machine Learning Approach for Medical Diagnosis Based on Prediction Model
Hemant Kasturiwale1*, Rajesh Karhe2 and Sujata N. Kale3
1Thakur College of Engineering and Technology, Kandivali (East), Mumbai, MS, India
2Shri Gulabrao Deokar College of Engineering, Jalgaon, MS, India
3Department of Applied Electronics, Sant Gadge Baba University, Amravati, MS, India
Abstract
The electrocardiography is the most crucial biosignals for critical analysis of the heart. The heart is the human body's most vital and variety of control mechanisms that regulate the heart's activities. The heart rate is an essential measure of cardiac function. The heart rate is represented as a time interval equal between two corresponding electrocardiogram (ECG) "R" peaks. The heart rate varies with the heart's state. A machine learning technique is used to categorize the statistical parameters mentioned above to predict the individual's physical state, including sleep, examination, and exercise, based on a physiologically important factor known as HRV. The chapter is focused on uses of manual classified data. Each hospital, clinic, and diagnostic center produces massive quantities of information such as patient records and test results to predict the presence of heart disease and provide care for the early stages. The results are validated and compared with predictions obtained from different algorithms. Classification and prediction are a mining technique that uses training data to construct a model, and then, that model is applied to test data to predict outcomes. Different algorithms are employed to disease datasets to diagnose chronic disease, and the findings have been positive. There is a need to establish an appropriate technique for the diagnosis of chronic diseases. This chapter discusses with insight various kinds of classification schemes for chronic disease prediction. Here, readers will come to choice know machine learning and classifiers made to get knowledge out of datasets.
Keywords: ECG, biosignals, machine learning, HRV, classification, prediction, cardiac diseases
1.1 Introduction
Biosignals are being used in various medical data, such as the electroencephalography (EEG), capturing electric fields created by brain cell activity, and magnetoencephalography (MEG) capturing magnet fields produced by electrical brain cell activity. The electrical stimulation comes from biological activity in various parts of the body. The most popular types of methods currently used to record biosignals in clinical research are described below, along with a brief overview of their functionality and related clinical application signals [1].
1.1.1 Heart System and Major Cardiac Diseases
The electrical activity generates the following types of signals:
- Magnetoencephalography (MEG) signals
- Electromyography (EMG) signals
- Electrooculography (EOG)signals
- Phonocardiography (PCG) signals
- Electrocorticography (ECoG) signals
- Electrocardiography (ECG or EKG) signals
Intervals between the waves are used as indicators of irregular cardiac operation, e.g., a prolonged PR interval from atrial activation to the start of ventricular activation may indicate cardiac failure [2, 3]. In addition, ECGs are used to study arrhythmias [4], coronary artery disease [5], and other heart failure disorders. In biosignals, the sampling frequency (or sampling rate) and the recording period are directly proportional to the data size and the data acquisition process speed. The ECG will be essential for the heart rhythm and disease research. The different heart conditions are as follows:
- a) Arrhythmias
- b) Coronary heart disease
- c) Various types of heart blocks
- d) Fibrillations
- e) Congestive heart failure (CHF)
- f) Myocardial infarction (MI)
- g) Premature ventricular contraction (PVC)
1.1.2 ECG for Heart Rate Variability Analysis
Electrocardiogram (ECG) is a waveform pattern that describes the state of cardiac activity and cardiac safety. The ECG signal is non-stationary and non-linear. The ECG has a spectrum of frequencies between 0.05 and 100 Hz [6]. ECG analysis methods, including the heart rate variability (HRV), QRS identification, and ECG post-processing, have advanced considerably since device implementation. The word HRV reflects the interval difference between successive heartbeats.
1.1.3 HRV for Cardiac Analysis
The biomedical signal is an important health assessment parameter. For example, it has been used to detect and predict human stress [1], stroke, hypertension, sleep disorder, age, gender, and many more. The popular techniques to analyze the HRV fall into three categories as time domain, spectral or frequency domain based on fast Fourier transform (FFT) [7], and nonlinear methods consisting of Markov modeling, entropy-based metrics [8], and probabilistic modeling [9]. There are seven commonly used statistical time domain parameters [10] calculated from HRV segmentation during 5-min recording, comprising of RMSSD, SDNN, SDANN, SDANNi, SDSD, PNN50, and autocorrelation, which are considered for implementation. The HRV is also calculated by a device called PPA (peripheral pulse analyzer); it works based on pulses measured, which is different from HRV measurement using ECG. However, the focus would be on ECG-based HRV measurement, but the validation PPA-based method is considered [11]. Nonlinear measurement approaches aim to calculate the structure and complexity of the time series of RR intervals. HRV signals are non-stationary and nonlinear in nature. Analysis of HRV dynamics by methods based on chaos theory and nonlinear system theory is based on findings indicating that the processes involved in cardiovascular control are likely to interact with each other in a nonlinear manner. The more on indices (features/parameters) are discussed in Section 1.3.2.
1.2 Machine Learning Approach and Prediction
Learning is closely connected to (and sometimes overlaps with) quantitative statistics, which often concentrate on forecasting computers' use. It has close connections with mathematical optimization, which provides the fields of methodology, theory, and implementation. The second sub-area focuses more on the study of exploratory data and is also known as non-monitored learning [2]. Unsupervised machine learning (ML) is also possible [11] and can be used to learn and construct baseline conduct profiles for different entities [12]. To gain knowledge of the past and to detect useful trends from massive, unstructured, and complex databases, machine learning algorithms use a range of statistical, probabilistic, and optimization methods [12]. These algorithms include automatic categorization of texts, network intrusion detection, junk e-mail filtering, credit-card fraud detection, consumer buying behavior, manufacturing optimization, and disease modeling. Most of these applications are performed using managed variants of the algorithms of ML rather than unattended [13].
The heart disease detail includes several features that predict heart disease. This large amount of medical data allowed data mining techniques to discover trends and diagnose patients. The historical medical data is very high, so it requires computational methods to process it. Data mining is a technique that removes the hidden pattern and uses as an analytical tool to analyze historical data. There are several different classification schemes for disease datasets. ML techniques are applied for classifying the statistical parameters above in a cardiological signal analysis to predict the RR interval estimate cannot be overemphasized. A precise method of calculation therefore needs to be developed. It is clear from the existing research theory that the conventional systems for chronic disease prediction are unable to establish reliable diagnostic systems as workers make it difficult to get correct responses and can minimize response time. Adaptive systems, by comparison, can increase the chances of success and can advise clinicians on care decisions. Current healthcare programmers can be enhanced by the efficient use of parallel classification systems, as they promote parallel implementation on multiple systems. Parallel classification systems also have a great potential to increase the predictive performance of diagnostic systems for chronic diseases [13, 14]. Here, classifiers are discussed out of the available are K-Nearest Neighbor (KNN), Support Vector Machine (SVM), Ensemble AdaBoost (EAB), and Random Forest (RF).
1.3 Material and Experimentation
The proposed method comprises of two phases:
- processing the enrolment database (PEP) and
- Prediction (P).
Figure 1.1 shows that the research purpose types of database are created based on acquisition units. The standard database has varying sampling frequency which comprises of different age groups of male and female.
A total number of subjects and corresponding signal were acquired with different set conditions. This may comprises of female and male with varying age group with sampling frequency of 256 and 500 Hz [6, 15]. The model will be testing for cardiac HRV-based analysis with both the ECG and non-ECG (PPA). For the research purpose, the congestive heart failure, arrhythmia, sudden cardiac death, ventricular arrhythmia, CHF database data being considered along with externally obtained ECG and non-ECG.
