Details

<p>Preface xv</p> <p><b>1 Probabilistic Optimization of Machine Learning Algorithms for Heart Disease Prediction 1<br /></b><i>Jaspreet Kaur, Bharti Joshi and Rajashree Shedge</i></p> <p>1.1 Introduction 2</p> <p>1.1.1 Scope and Motivation 3</p> <p>1.2 Literature Review 4</p> <p>1.2.1 Comparative Analysis 5</p> <p>1.2.2 Survey Analysis 5</p> <p>1.3 Tools and Techniques 10</p> <p>1.3.1 Description of Dataset 11</p> <p>1.3.2 Machine Learning Algorithm 12</p> <p>1.3.3 Decision Tree 14</p> <p>1.3.4 Random Forest 15</p> <p>1.3.5 Naive Bayes Algorithm 16</p> <p>1.3.6 K Means Algorithm 18</p> <p>1.3.7 Ensemble Method 18</p> <p>1.3.7.1 Bagging 19</p> <p>1.3.7.2 Boosting 19</p> <p>1.3.7.3 Stacking 19</p> <p>1.3.7.4 Majority Vote 19</p> <p>1.4 Proposed Method 20</p> <p>1.4.1 Experiment and Analysis 20</p> <p>1.4.2 Method 22</p> <p>1.5 Conclusion 25</p> <p>References 26</p> <p><b>2 Cancerous Cells Detection in Lung Organs of Human Body: IoT-Based Healthcare 4.0 Approach 29<br /></b><i>Rohit Rastogi, D.K. Chaturvedi, Sheelu Sagar, Neeti Tandon and Mukund Rastogi</i></p> <p>2.1 Introduction 30</p> <p>2.1.1 Motivation to the Study 30</p> <p>2.1.1.1 Problem Statements 31</p> <p>2.1.1.2 Authors’ Contributions 31</p> <p>2.1.1.3 Research Manuscript Organization 31</p> <p>2.1.1.4 Definitions 32</p> <p>2.1.2 Computer-Aided Diagnosis System (CADe or CADx) 32</p> <p>2.1.3 Sensors for the Internet of Things 32</p> <p>2.1.4 Wireless and Wearable Sensors for Health Informatics 33</p> <p>2.1.5 Remote Human’s Health and Activity Monitoring 33</p> <p>2.1.6 Decision-Making Systems for Sensor Data 33</p> <p>2.1.7 Artificial Intelligence and Machine Learning for Health Informatics 34</p> <p>2.1.8 Health Sensor Data Management 34</p> <p>2.1.9 Multimodal Data Fusion for Healthcare 35</p> <p>2.1.10 Heterogeneous Data Fusion and Context-Aware Systems: A Context-Aware Data Fusion Approach for Health-IoT 35</p> <p>2.2 Literature Review 35</p> <p>2.3 Proposed Systems 37</p> <p>2.3.1 Framework or Architecture of the Work 38</p> <p>2.3.2 Model Steps and Parameters 38</p> <p>2.3.3 Discussions 39</p> <p>2.4 Experimental Results and Analysis 39</p> <p>2.4.1 Tissue Characterization and Risk Stratification 39</p> <p>2.4.2 Samples of Cancer Data and Analysis 40</p> <p>2.5 Novelties 42</p> <p>2.6 Future Scope, Limitations, and Possible Applications 42</p> <p>2.7 Recommendations and Consideration 43</p> <p>2.8 Conclusions 43</p> <p>References 43</p> <p><b>3 Computational Predictors of the Predominant Protein Function: SARS-CoV-2 Case 47<br /></b><i>Carlos Polanco, Manlio F. Márquez and Gilberto Vargas-Alarcón</i></p> <p>3.1 Introduction 48</p> <p>3.2 Human Coronavirus Types 49</p> <p>3.3 The SARS-CoV-2 Pandemic Impact 50</p> <p>3.3.1 RNA Virus vs DNA Virus 51</p> <p>3.3.2 The <i>Coronaviridae </i>Family 51</p> <p>3.3.3 The SARS-CoV-2 Structural Proteins 52</p> <p>3.3.4 Protein Representations 52</p> <p>3.4 Computational Predictors 53</p> <p>3.4.1 Supervised Algorithms 53</p> <p>3.4.2 Non-Supervised Algorithms 54</p> <p>3.5 Polarity Index Method^® 54</p> <p>3.5.1 The PIM^® Profile 54</p> <p>3.5.2 Advantages 55</p> <p>3.5.3 Disadvantages 55</p> <p>3.5.4 SARS-CoV-2 Recognition Using PIM^® Profile 55</p> <p>3.6 Future Implications 59</p> <p>3.7 Acknowledgments 60</p> <p>References 60</p> <p><b>4 Deep Learning in Gait Abnormality Detection: Principles and Illustrations 63<br /></b><i>Saikat Chakraborty, Sruti Sambhavi and Anup Nandy</i></p> <p>4.1 Introduction 63</p> <p>4.2 Background 65</p> <p>4.2.1 LSTM 65</p> <p>4.2.1.1 Vanilla LSTM 65</p> <p>4.2.1.2 Bidirectional LSTM 66</p> <p>4.3 Related Works 67</p> <p>4.4 Methods 68</p> <p>4.4.1 Data Collection and Analysis 68</p> <p>4.4.2 Results and Discussion 69</p> <p>4.5 Conclusion and Future Work 71</p> <p>4.6 Acknowledgments 71</p> <p>References 71</p> <p><b>5 Broad Applications of Network Embeddings in Computational Biology, Genomics, Medicine, and Health 73<br /></b><i>Akanksha Jaiswar, Devender Arora, Manisha Malhotra, Abhimati Shukla and Nivedita Rai</i></p> <p>5.1 Introduction 74</p> <p>5.2 Types of Biological Networks 76</p> <p>5.3 Methodologies in Network Embedding 76</p> <p>5.4 Attributed and Non-Attributed Network Embedding 82</p> <p>5.5 Applications of Network Embedding in Computational Biology 83</p> <p>5.5.1 Understanding Genomic and Protein Interaction via Network Alignment 83</p> <p>5.5.2 Pharmacogenomics 84</p> <p>5.5.2.1 Drug-Target Interaction Prediction 84</p> <p>5.5.2.2 Drug-Drug Interaction 84</p> <p>5.5.2.3 Drug-Disease Interaction Prediction 85</p> <p>5.5.2.4 Analysis of Adverse Drug Reaction 85</p> <p>5.5.3 Function Prediction 86</p> <p>5.5.4 Community Detection 86</p> <p>5.5.5 Network Denoising 87</p> <p>5.5.6 Analysis of Multi-Omics Data 87</p> <p>5.6 Limitations of Network Embedding in Biology 87</p> <p>5.7 Conclusion and Outlook 89</p> <p>References 89</p> <p><b>6 Heart Disease Classification Using Regional Wall Thickness by Ensemble Classifier 99<br /></b><i>Prakash J., Vinoth Kumar B. and Sandhya R.</i></p> <p>6.1 Introduction 100</p> <p>6.2 Related Study 101</p> <p>6.3 Methodology 103</p> <p>6.3.1 Pre-Processing 103</p> <p>6.3.2 Region of Interest Extraction 104</p> <p>6.3.3 Segmentation 105</p> <p>6.3.4 Feature Extraction 106</p> <p>6.3.5 Disease Classification 107</p> <p>6.4 Implementation and Result Analysis 108</p> <p>6.4.1 Dataset Description 108</p> <p>6.4.2 Testbed 108</p> <p>6.4.3 Discussion 108</p> <p>6.4.3.1 K-Fold Cross-Validation 110</p> <p>6.4.3.2 Confusion Matrix 110</p> <p>6.5 Conclusion 115</p> <p>References 115</p> <p><b>7 Deep Learning for Medical Informatics and Public Health 117<br /></b><i>K. Aditya Shastry, Sanjay H. A., Lakshmi M. and Preetham N.</i></p> <p>7.1 Introduction 118</p> <p>7.2 Deep Learning Techniques in Medical Informatics and Public Health 121</p> <p>7.2.1 Autoencoders 122</p> <p>7.2.2 Recurrent Neural Network 123</p> <p>7.2.3 Convolutional Neural Network (CNN) 124</p> <p>7.2.4 Deep Boltzmann Machine 126</p> <p>7.2.5 Deep Belief Network 127</p> <p>7.3 Applications of Deep Learning in Medical Informatics and Public Health 128</p> <p>7.3.1 The Use of DL for Cancer Diagnosis 128</p> <p>7.3.2 DL in Disease Prediction and Treatment 129</p> <p>7.3.3 Future Applications 133</p> <p>7.4 Open Issues Concerning DL in Medical Informatics and Public Health 135</p> <p>7.5 Conclusion 139</p> <p>References 140</p> <p><b>8 An Insight Into Human Pose Estimation and Its Applications 147<br /></b><i>Shambhavi Mishra, Janamejaya Channegowda and Kasina Jyothi Swaroop</i></p> <p>8.1 Foundations of Human Pose Estimation 147</p> <p>8.2 Challenges to Human Pose Estimation 149</p> <p>8.2.1 Motion Blur 150</p> <p>8.2.2 Indistinct Background 151</p> <p>8.2.3 Occlusion or Self-Occlusion 151</p> <p>8.2.4 Lighting Conditions 151</p> <p>8.3 Analyzing the Dimensions 152</p> <p>8.3.1 2D Human Pose Estimation 152</p> <p>8.3.1.1 Single-Person Pose Estimation 153</p> <p>8.3.1.2 Multi-Person Pose Estimation 153</p> <p>8.3.2 3D Human Pose Estimation 153</p> <p>8.4 Standard Datasets for Human Pose Estimation 154</p> <p>8.4.1 Pascal VOC (Visual Object Classes) Dataset 156</p> <p>8.4.2 KTH Multi-View Football Dataset I 156</p> <p>8.4.3 KTH Multi-View Football Dataset II 156</p> <p>8.4.4 MPII Human Pose Dataset 157</p> <p>8.4.5 BBC Pose 157</p> <p>8.4.6 COCO Dataset 157</p> <p>8.4.7 J-HMDB Dataset 158</p> <p>8.4.8 Human3.6M Dataset 158</p> <p>8.4.9 DensePose 158</p> <p>8.4.10 AMASS Dataset 159</p> <p>8.5 Deep Learning Revolutionizing Pose Estimation 159</p> <p>8.5.1 Approaches in 2D Human Pose Estimation 159</p> <p>8.5.2 Approaches in 3D Human Pose Estimation 163</p> <p>8.6 Application of Human Pose Estimation in Medical Domains 165</p> <p>8.7 Conclusion 166</p> <p>References 167</p> <p><b>9 Brain Tumor Analysis Using Deep Learning: Sensor and IoT-Based Approach for Futuristic Healthcare 171<br /></b><i>Rohit Rastogi, D.K. Chaturvedi, Sheelu Sagar, Neeti Tandon and Akshit Rajan Rastogi</i></p> <p>9.1 Introduction 172</p> <p>9.1.1 Brain Tumor 172</p> <p>9.1.2 Big Data Analytics in Health Informatics 172</p> <p>9.1.3 Machine Learning in Healthcare 173</p> <p>9.1.4 Sensors for Internet of Things 173</p> <p>9.1.5 Challenges and Critical Issues of IoT in Healthcare 174</p> <p>9.1.6 Machine Learning and Artificial Intelligence for Health Informatics 174</p> <p>9.1.7 Health Sensor Data Management 175</p> <p>9.1.8 Multimodal Data Fusion for Healthcare 175</p> <p>9.1.9 Heterogeneous Data Fusion and Context-Aware Systems a Context-Aware Data Fusion Approach for Health-IoT 176</p> <p>9.1.10 Role of Technology in Addressing the Problem of Integration of Healthcare System 176</p> <p>9.2 Literature Survey 177</p> <p>9.3 System Design and Methodology 179</p> <p>9.3.1 System Design 179</p> <p>9.3.2 CNN Architecture 180</p> <p>9.3.3 Block Diagram 181</p> <p>9.3.4 Algorithm(s) 181</p> <p>9.3.5 Our Experimental Results, Interpretation, and Discussion 183</p> <p>9.3.6 Implementation Details 183</p> <p>9.3.7 Snapshots of Interfaces 184</p> <p>9.3.8 Performance Evaluation 186</p> <p>9.3.9 Comparison with Other Algorithms 186</p> <p>9.4 Novelty in Our Work 186</p> <p>9.5 Future Scope, Possible Applications, and Limitations 188</p> <p>9.6 Recommendations and Consideration 188</p> <p>9.7 Conclusions 188</p> <p>References 189</p> <p><b>10 Study of Emission From Medicinal Woods to Curb Threats of Pollution and Diseases: Global Healthcare Paradigm Shift in 21st Century 191<br /></b><i>Rohit Rastogi, Mamta Saxena, Devendra Kr. Chaturvedi, Sheelu Sagar, Neha Gupta, Harshit Gupta, Akshit Rajan Rastogi, Divya Sharma, Manu Bhardwaj and Pranav Sharma</i></p> <p>10.1 Introduction 192</p> <p>10.1.1 Scenario of Pollution and Need to Connect with Indian Culture 192</p> <p>10.1.2 Global Pollution Scenario 192</p> <p>10.1.3 Indian Crisis on Pollution and Worrying Stats 193</p> <p>10.1.4 Efforts Made to Curb Pollution World Wide 194</p> <p>10.1.5 Indian Ancient Vedic Sciences to Curb Pollution and Related Disease 196</p> <p>10.1.6 The Yajna Science: A Boon to Human Race From Rishi-Muni 196</p> <p>10.1.7 The Science of Mantra Associated With Yajna and Its Scientific Effects 197</p> <p>10.1.8 Effect of Different Woods and Cow Dung Used in Yajna 197</p> <p>10.1.9 Use of Sensors and IoT to Record Experimental Data 198</p> <p>10.1.10 Analysis and Pattern Recognition by ML and AI 199</p> <p>10.2 Literature Survey 200</p> <p>10.3 The Methodology and Protocols Followed 201</p> <p>10.4 Experimental Setup of an Experiment 202</p> <p>10.5 Results and Discussions 202</p> <p>10.5.1 Mango 202</p> <p>10.5.2 Bargad 203</p> <p>10.6 Applications of Yagya and Mantra Therapy in Pollution Control and Its Significance 207</p> <p>10.7 Future Research Perspectives 207</p> <p>10.8 Novelty of Our Research 208</p> <p>10.9 Recommendations 208</p> <p>10.10 Conclusions 209</p> <p>References 209</p> <p><b>11 An Economical Machine Learning Approach for Anomaly Detection in IoT Environment 215<br /></b><i>Ambika N.</i></p> <p>11.1 Introduction 215</p> <p>11.2 Literature Survey 218</p> <p>11.3 Proposed Work 229</p> <p>11.4 Analysis of the Work 230</p> <p>11.5 Conclusion 231</p> <p>References 231</p> <p><b>12 Indian Science of Yajna and Mantra to Cure Different Diseases: An Analysis Amidst Pandemic With a Simulated Approach 235<br /></b><i>Rohit Rastogi, Mamta Saxena, Devendra Kumar Chaturvedi, Mayank Gupta, Puru Jain, Rishabh Jain, Mohit Jain, Vishal Sharma, Utkarsh Sangam, Parul Singhal and Priyanshi Garg</i></p> <p>12.1 Introduction 236</p> <p>12.1.1 Different Types of Diseases 236</p> <p>12.1.1.1 Diabetes (Madhumeha) and Its Types 236</p> <p>12.1.1.2 TTH and Stress 237</p> <p>12.1.1.3 Anxiety 237</p> <p>12.1.1.4 Hypertension 237</p> <p>12.1.2 Machine Vision 237</p> <p>12.1.2.1 Medical Images and Analysis 238</p> <p>12.1.2.2 Machine Learning in Healthcare 238</p> <p>12.1.2.3 Artificial Intelligence in Healthcare 239</p> <p>12.1.3 Big Data and Internet of Things (IoT) 239</p> <p>12.1.4 Machine Learning in Association with Data Science and Analytics 239</p> <p>12.1.5 Yajna Science 240</p> <p>12.1.6 Mantra Science 240</p> <p>12.1.6.1 Positive Impact of Recital of Gayatri Mantra and OM Chanting 241</p> <p>12.1.6.2 Significance of Mantra on Indian Culture and Mythology 241</p> <p>12.1.7 Usefulness and Positive Aspect of Yoga Asanas and Pranayama 241</p> <p>12.1.8 Effects of Yajna and Mantra on Human Health 242</p> <p>12.1.9 Impact of Yajna in Reducing the Atmospheric Solution 242</p> <p>12.1.10 Scientific Study on Impact of Yajna on Air Purification 243</p> <p>12.1.11 Scientific Meaning of Religious and Manglik Signs 244</p> <p>12.2 Literature Survey 244</p> <p>12.3 Methodology 246</p> <p>12.4 Results and Discussion 249</p> <p>12.5 Interpretations and Analysis 250</p> <p>12.6 Novelty in Our Work 258</p> <p>12.7 Recommendations 259</p> <p>12.8 Future Scope and Possible Applications 260</p> <p>12.9 Limitations 261</p> <p>12.10 Conclusions 261</p> <p>12.11 Acknowledgments 262</p> <p>References 262</p> <p><b>13 Collection and Analysis of Big Data From Emerging Technologies in Healthcare 269<br /></b><i>Nagashri K., Jayalakshmi D. S. and Geetha J.</i></p> <p>13.1 Introduction 269</p> <p>13.2 Data Collection 271</p> <p>13.2.1 Emerging Technologies in Healthcare and Its Applications 271</p> <p>13.2.1.1 RFID 272</p> <p>13.2.1.2 WSN 273</p> <p>13.2.1.3 IoT 274</p> <p>13.2.2 Issues and Challenges in Data Collection 277</p> <p>13.2.2.1 Data Quality 277</p> <p>13.2.2.2 Data Quantity 277</p> <p>13.2.2.3 Data Access 278</p> <p>13.2.2.4 Data Provenance 278</p> <p>13.2.2.5 Security 278</p> <p>13.2.2.6 Other Challenges 279</p> <p>13.3 Data Analysis 280</p> <p>13.3.1 Data Analysis Approaches 280</p> <p>13.3.1.1 Machine Learning 280</p> <p>13.3.1.2 Deep Learning 281</p> <p>13.3.1.3 Natural Language Processing 281</p> <p>13.3.1.4 High-Performance Computing 281</p> <p>13.3.1.5 Edge-Fog Computing 282</p> <p>13.3.1.6 Real-Time Analytics 282</p> <p>13.3.1.7 End-User Driven Analytics 282</p> <p>13.3.1.8 Knowledge-Based Analytics 283</p> <p>13.3.2 Issues and Challenges in Data Analysis 283</p> <p>13.3.2.1 Multi-Modal Data 283</p> <p>13.3.2.2 Complex Domain Knowledge 283</p> <p>13.3.2.3 Highly Competent End-Users 283</p> <p>13.3.2.4 Supporting Complex Decisions 283</p> <p>13.3.2.5 Privacy 284</p> <p>13.3.2.6 Other Challenges 284</p> <p>13.4 Research Trends 284</p> <p>13.5 Conclusion 286</p> <p>References 286</p> <p><b>14 A Complete Overview of Sign Language Recognition and Translation Systems 289<br /></b><i>Kasina Jyothi Swaroop, Janamejaya Channegowda and Shambhavi Mishra</i></p> <p>14.1 Introduction 289</p> <p>14.2 Sign Language Recognition 290</p> <p>14.2.1 Fundamentals of Sign Language Recognition 290</p> <p>14.2.2 Requirements for the Sign Language Recognition 292</p> <p>14.3 Dataset Creation 293</p> <p>14.3.1 American Sign Language 293</p> <p>14.3.2 German Sign Language 296</p> <p>14.3.3 Arabic Sign Language 297</p> <p>14.3.4 Indian Sign Language 298</p> <p>14.4 Hardware Employed for Sign Language Recognition 299</p> <p>14.4.1 Glove/Sensor-Based Systems 299</p> <p>14.4.2 Microsoft Kinect–Based Systems 300</p> <p>14.5 Computer Vision–Based Sign Language Recognition and Translation Systems 302</p> <p>14.5.1 Image Processing Techniques for Sign Language Recognition 302</p> <p>14.5.2 Deep Learning Methods for Sign Language Recognition 304</p> <p>14.5.3 Pose Estimation Application to Sign Language Recognition 305</p> <p>14.5.4 Temporal Information in Sign Language Recognition and Translation 306</p> <p>14.6 Sign Language Translation System—A Brief Overview 307</p> <p>14.7 Conclusion 309</p> <p>References 310</p> <p>Index 315</p>

<p><b>A. Suresh, PhD</B> is an associate professor, Department of the Networking and Communications, SRM Institute of Science & Technology, Kattankulathur, Tamil Nadu, India. He has nearly two decades of experience in teaching and his areas of specialization are data mining, artificial intelligence, image processing, multimedia, and system software. He has published 6 patents and more than 100 papers in international journals.</p> <p><b>S. Vimal, PhD</B> is an assistant professor in the Department of Artificial Intelligence & DS, Ramco Institute of Technology, Tamilnadu, India. He is the editor of 3 books and guest-edited multiple journal special issues. He has more than 15 years of teaching experience. <p><b>Y. Harold Robinson, PhD </B>is currently working in the School of Technology and Engineering, Vellore Institute of Technology, Vellore, India. He has published more than 50 papers in various international journals and presented more than 70 papers in both national and international conferences. <p><b>Dhinesh Kumar Ramaswami, </B>BE in Computer Science, is a Senior Consultant at Capgemini America Inc. He has over 9 years of experience in software development and specializes in various .net technologies. He has published more than 15 papers in international journals and national and international conferences. <p><b>R. Udendhran, PhD </B>is an assistant professor, Department of Computer Science and Engineering at Sri Sairam Institute of Technology, Chennai, Tamil Nadu, India. He has published about 20 papers in international journals.

<p><b>The main topics addressed in this book are big data analytics problems in bioinformatics research such as microarray data analysis, sequence analysis, genomics-based analytics, disease network analysis, techniques for big data analytics, and health information technology.</b></p> <p><i>Bioinformatics and Medical Applications: Big Data Using Deep Learning Algorithms</i> analyses massive biological datasets using computational approaches and the latest cutting-edge technologies to capture and interpret biological data. The book delivers various bioinformatics computational methods used to identify diseases at an early stage by assembling cutting-edge resources into a single collection designed to enlighten the reader on topics focusing on computer science, mathematics, and biology. In modern biology and medicine, bioinformatics is critical for data management. This book explains the bioinformatician’s important tools and examines how they are used to evaluate biological data and advance disease knowledge. <p>The editors have curated a distinguished group of perceptive and concise chapters that presents the current state of medical treatments and systems and offers emerging solutions for a more personalized approach to healthcare. Applying deep learning techniques for data-driven solutions in health information allows automated analysis whose method can be more advantageous in supporting the problems arising from medical and health-related information. <p><b>Audience</b> <p>The primary audience for the book includes specialists, researchers, postgraduates, designers, experts, and engineers, who are occupied with biometric research and security-related issues.

US