Decision Intelligence: Proceedings of the International Conference on Information Technology, InCITe 2023, Volume 1 (Lecture Notes in Electrical Engineering, 1079) 9819959969, 9789819959969

Table of contents :
Preface
Contents
About the Editors
A Radio Frequency-Based Energy Harvesting Model for IoMT Device
1 Introduction
1.1 Contribution
2 Related Works
3 Proposed Model
3.1 RFEH Model
3.2 RFEH-IoMT Model
4 Performance Analysis
5 Conclusion
References
Cancer Hotspot Identification and Analysis: A Scan Statistics Approach
1 Introduction
1.1 Research Problem
1.2 Related Work
2 Relevance
3 Objective
4 Data Design and Processing
5 Methodology Adopted
5.1 Hotspot Analysis
5.2 Welch’s T-test
6 Spearman’s Rank Correlation Coefficient (rs)
7 Results and Discussion
7.1 Hotspot Detection
7.2 Descriptive Approach
8 Conclusion
9 Limitations
10 Scope for Future Research
References
An Artificial Voice Box that Makes Use of Unconventional Methods of Machine Learning
1 Introduction
2 Related Work
3 Materials and Methods
3.1 Neuro Decoder
3.2 Artificial Voice Box
4 Conclusion
References
State-of-Art Review on Medical Image Classification Techniques
1 Introduction
2 Image Pre-processing Techniques for Medical Images
2.1 Image Filtering
2.2 Segmentation
3 Feature Selection and Feature Extraction
4 Classification Techniques
4.1 Using Traditional Classifiers
4.2 Using Deep Learning Algorithms
5 Conclusion
References
Prediction of Loan Approval of Customers Based on Credit Score Using Machine Learning
1 Introduction
2 Motivation
3 Literature Survey
4 Methodology
4.1 Dataset
4.2 Credit Score
4.3 Pre-processing
4.4 Logistic Regression
4.5 K-Nearest Neighbour
4.6 Decision Tree
4.7 Random Forest
4.8 Linear Discriminant Analysis
5 Result Analysis
6 Conclusion and Future Scope
References
Strategies for the Adoption of AI Technologies in the South African Wine and Fruit Industries
1 Introduction
2 Literature Review
2.1 AI Technologies for the SA Agricultural Sector
2.2 Barriers and Challenges of Adopting AI Technologies in Agriculture
3 Research Design and Methodology
4 Findings and Discussion
4.1 Current AI Technologies in Agricultural Processes
4.2 Challenges of Adopting AI Technologies
4.3 Strategies to Facilitate the Adoption of AI Technologies in Agriculture
5 Conclusion
References
Real-Time Facial Mask Detection Using Deep Learning
1 Introduction
2 Related Work
3 Methodology
3.1 Data Collection
3.2 Model
3.3 Training
4 Result
5 Limitations
6 Conclusion and Future Scope
References
SSATS—Enhancement of Semantic Similarity of Abstractive Text Summarization Using Transformer
1 Introduction
2 Key Contributions
3 Literature Survey
4 Proposed Methodology
5 Experiments
5.1 Dataset
5.2 Metrics
6 Results and Discussion
7 Conclusion and Future Work
References
Semantic Segmentation of Optical Satellite Images
1 Introduction
2 Evaluation Metric
3 Related Work
4 Comparison Between State-of-the-Art Methods
5 Future Work and Challenges
6 Conclusion
References
Human Gender and Age Estimator Using Local and Global Features with Machine Learning
1 Introduction
2 Literature Survey
3 Methodology
3.1 Data Acquisition
3.2 Pre-processing
3.3 Face Detection
3.4 Adaboost Algorithm
3.5 Feature Extraction
3.6 HOG Descriptors—Histogram of Oriented Gradients
3.7 LBP—Local Binary Pattern
3.8 Feature Reduction Using PCA
3.9 Principle Component Analysis (PCA)
3.10 SVM—Support Vector Machine Classifier
4 Experimental Results
4.1 Training and Testing
5 Conclusion
References
Predictive Analysis of Neurodegenerative and Chronic Fatigue Disease
1 Introduction
1.1 Literature Review
2 Methodology
References
An Automatic Early Alert System on Detecting Dozing Driver
1 Introduction
2 Related Research Work
3 Proposed Methodology
3.1 Facial Landmark Identification
3.2 Eye Closing Detection
3.3 Yawning Detection
3.4 Alert
4 Experimental Results
5 Conclusion
References
Reversible Image Steganography to Achieve Effective PSNR
1 Introduction
2 Literature Review
3 Methodology
3.1 Reversible Image Steganography Methods for Compressed Images
3.2 Reversible Steganography Methods for Coded Images
3.3 High-Capacity Techniques for Reversible Image Steganography
3.4 Performance Study for Reversible Image Steganography
4 Experimental Results
5 Conclusion
References
EDA and Predicting Customer’s Response for Cross-Sell Vehicle Insurance
1 Introduction
2 Literature Review
2.1 Why ML in Insurance
2.2 Related Work
3 Data Collection and Understanding
4 Work Flow and Framework of the Proposed Analysis
5 Data Preparation
6 Techniques Employed
7 Results and Discussion
8 Conclusion and Future Work
References
A Study of Feature-Based and Pixel-Level Image Fusion Techniques
1 Introduction
2 Pixel Fusion
3 Feature Level
4 Discrete Wavelet Transform
5 C-Mean Fuzzy Clustering
6 The Proposed Model
7 Performance Measures
7.1 Standard Deviation (SD)
7.2 Entropy (En)
7.3 Signal-to-Noise Ratio (SNR)
7.4 Deviation Index (DI)
8 Results
9 Conclusion
References
Catch Your Session, Track Your Pulse e-Health Service Using Blockchain
1 Introduction
2 Literature Survey
3 Methodology
4 Proposed System
5 Architecture System
5.1 Patient information management Module
5.2 Slot Booking Module
6 Results and Analysis
7 Conclusion
References
Ansuni Baat—A Bridge to Sign Language Communication
1 Introduction
1.1 Aim and Features
1.2 Background of the Study
2 Literature Review
3 System Analysis
3.1 Software Requirement Specification
3.2 Flow Diagram for the System
4 Methodology
4.1 Front-end
4.2 Back-end
4.3 Database
5 Result
6 Conclusion
7 Future Scope
References
A Review of Deep Learning Algorithms for Early Detection of Oral Mouth Cancer
1 Introduction
2 Research Approach
2.1 Research Question
2.2 Data Source
2.3 Paper Selection Criteria
3 Findings
4 Future Challenges
5 Conclusion
References
Malware Detection Using Deep Learning
1 Literature Review
2 Linear Regression
3 Types of Malwares
3.1 Virus
3.2 Worm
3.3 Trojan Horse
3.4 Spyware
3.5 Rootkit
4 Malware Detection
4.1 Anomaly Detection
4.2 Signature Detection
4.3 Specification-Based Detection
5 Botnet
5.1 How Botnet Attacks IoT Devices
5.2 Feature Selection
6 Algorithms Used
6.1 Logistic Regression
6.2 F1 Score
6.3 Epoch
6.4 KERAS
6.5 Standardization
7 Attributes of Datasets Used in the Project
7.1 KNN
7.2 Decision Tree
7.3 ANN
7.4 Logistic Regression
7.5 Naïve Bayes
7.6 SVM
8 Conclusion
References
Automatic Title Generation with Attention-Based LSTM
1 Introduction
1.1 Motivation and Justification of Automatic Title Generation
1.2 Contribution of the Proposed Work
2 Related Work
3 Proposed Methodology
3.1 Sequence-to-Sequence (Seq-to-Seq) Model
3.2 Seq-to-Seq Model Components
4 Experimental Setup and Result
4.1 Data Collection
4.2 Data Preprocessing
4.3 Train the Processed Data Using the Seq-to-Seq Model
5 Conclusion of the Proposed Work and Future Work
References
COVID-19 (COV-19) Spreading Diagnoses by Feature Representation Method Through Visual Learning (FVisL-CoV19)
1 Introduction
2 Related Works
3 Proposed Model
3.1 Feature Illustration of Visible-Learning-Methodology (FVisL-CoV19)
3.2 CoV19 Spreading Diagnoses Algorithm
4 Results and Discussions
4.1 Dataset Evaluation
4.2 Evaluation Metrics
4.3 Results
5 Conclusions
References
Alzheimer’s Disease Classification Using Deep Learning Models
1 Introduction
2 Literature Survey
3 Proposed Work
3.1 Input
3.2 Data Pre-processing
3.3 Model Building
4 Results and Discussion
5 Conclusion
References
Brain Pathology Detection Using Convolutional Neural Network from EEG Signal
1 Introduction
2 Related Works
3 Methodology and Design
4 Implementation and Results
5 Conclusion
References
CNN-Based Adversarial Embedding for Image Steganograpy
1 Introduction
2 Literature Review
3 Methods
3.1 Existing Methods
3.2 Proposed Methods
4 Result and Discussion
5 Conclusion
References
Machine Learning Approaches for Entity Extraction from Citation Strings
1 Introduction
2 Related Work
2.1 Non-machine Learning based Approaches
2.2 Machine Learning based Approaches
2.3 Limitations of Machine Learning based Approaches
2.4 Combination of Machine Learning and non-Machine Learning methods
3 Comparative Analysis
4 Conclusion
5 Future Scope
References
A Survey on Brain Tumor Segmentation with Missing MRI Modalities
1 Introduction
2 Methodologies
2.1 Common Latent Space
2.2 Knowledge Distillation Network
2.3 Mutual Information Maximization
3 Dataset
4 Evaluation Metrices
5 Experimental Analysis
6 Conclusion and Future Work
References
Deep Learning Based Segmentation Approach for Cervical Cancer Detection Using Pap Smears
1 Introduction
2 Literature Review
3 Methodology
3.1 Dataset
3.2 Data Augmentation
3.3 Data Pre-Processing
3.4 Data Augmentation
3.5 Segmentation
3.6 Performance Measures
3.7 Proposed Method
4 Result
4.1 Segmentation of Cervical Class
5 Conclusion
References
Hybrid Sign Language Interpreter Development Using Machine Learning Approach
1 Introduction
2 Literature Review
3 Research Gaps and Queries
4 Research Methodology
4.1 Implementation of the Methodology
4.2 Threats to Validity
5 Result Outcome
6 Conclusion and Future Scope
References
Community Detection Algorithms in Social Networks: An Empirical Evaluation
1 Introduction
2 Community Detection Algorithms and Measures
2.1 Community Detection Algorithms
2.2 Evaluation Measures
3 Results and Discussion
3.1 Datasets
3.2 Experimental Setup
3.3 Results
3.4 Discussion
4 Conclusion
References
Designing Hybrid Image Fusion Algorithm Using CNN and Stationary Wavelet Transform
1 Introduction
1.1 Spatial Based
1.2 Transform Based
2 Literature Review
3 Methodology
4 Experiments and Results
5 Conclusion
References
Predict Pawpularity Score of Pets Using State of Art Algorithms
1 Introduction
2 Literature Survey
3 Methodology
3.1 Data Exploration
3.2 Model Fitting
3.3 Models Used
4 Conclusion and Future Scope
References
Comparative Analysis of Balanced Code Smell Detection Using Machine Learning
1 Introduction
2 Literature Survey
3 Research Methodology
3.1 Dataset
4 Results and Discussion
5 Conclusions
References
ETL Data Pipeline to Analyze Scraped Data
1 Introduction
2 Literature Review
3 Proposed Method
4 Data Visualization in Power BI
References
DDoS Attack Detection Using Machine Learning
1 Introduction
2 Literature Review
3 Methodology
3.1 Dataset Description
3.2 Dataset Preprocessing
4 Results
5 Conclusion
References
An Empirical Study to Investigate Class Imbalance Issue for Improving Security Bug Report Classification Prediction
1 Introduction
2 Related Work
2.1 Security Bug Report Classifications
3 Methodology
3.1 Data-Preprocessing
3.2 Class Balancing Technique
3.3 Classifier Model
3.4 Performance Measure
4 Result and Analysis
5 Threats to Validity
6 Conclusion and Future Scope
References
Factors Influencing the Use of a Soccer Team’s Mobile Application by Fans in South Africa
1 Introduction
2 Literature Review
2.1 Mobile Applications in Sports
2.2 Impact of the Covid-19 Pandemic on the Sports Industry
2.3 Sport Fan Experience in Developed Versus Developing Countries
3 Research Model and Hypotheses
4 Research Design and Methodology
5 Data Analysis
5.1 Descriptive Statistics
5.2 Model Testing
6 Conclusion
References

Citation preview

Lecture Notes in Electrical Engineering 1079

B. K. Murthy B. V. R. Reddy Nitasha Hasteer Jean-Paul Van Belle   Editors

Decision Intelligence Proceedings of the International Conference on Information Technology, InCITe 2023, Volume 1

Lecture Notes in Electrical Engineering Volume 1079

Series Editors Leopoldo Angrisani, Department of Electrical and Information Technologies Engineering, University of Napoli Federico II, Napoli, Italy Marco Arteaga, Departament de Control y Robótica, Universidad Nacional Autónoma de México, Coyoacán, Mexico Samarjit Chakraborty, Fakultät für Elektrotechnik und Informationstechnik, TU München, München, Germany Jiming Chen, Zhejiang University, Hangzhou, Zhejiang, China Shanben Chen, School of Materials Science and Engineering, Shanghai Jiao Tong University, Shanghai, China Tan Kay Chen, Department of Electrical and Computer Engineering, National University of Singapore, Singapore, Singapore Rüdiger Dillmann, University of Karlsruhe (TH) IAIM, Karlsruhe, Baden-Württemberg, Germany Haibin Duan, Beijing University of Aeronautics and Astronautics, Beijing, China Gianluigi Ferrari, Dipartimento di Ingegneria dell’Informazione, Sede Scientifica Università degli Studi di Parma, Parma, Italy Manuel Ferre, Centre for Automation and Robotics CAR (UPM-CSIC), Universidad Politécnica de Madrid, Madrid, Spain Faryar Jabbari, Department of Mechanical and Aerospace Engineering, University of California, Irvine, CA, USA Limin Jia, State Key Laboratory of Rail Traffic Control and Safety, Beijing Jiaotong University, Beijing, China Janusz Kacprzyk, Intelligent Systems Laboratory, Systems Research Institute, Polish Academy of Sciences, Warsaw, Poland Alaa Khamis, Department of Mechatronics Engineering, German University in Egypt El Tagamoa El Khames, New Cairo City, Egypt Torsten Kroeger, Intrinsic Innovation, Mountain View, CA, USA Yong Li, College of Electrical and Information Engineering, Hunan University, Changsha, Hunan, China Qilian Liang, Department of Electrical Engineering, University of Texas at Arlington, Arlington, TX, USA Ferran Martín, Departament d’Enginyeria Electrònica, Universitat Autònoma de Barcelona, Bellaterra, Barcelona, Spain Tan Cher Ming, College of Engineering, Nanyang Technological University, Singapore, Singapore Wolfgang Minker, Institute of Information Technology, University of Ulm, Ulm, Germany Pradeep Misra, Department of Electrical Engineering, Wright State University, Dayton, OH, USA Subhas Mukhopadhyay, School of Engineering, Macquarie University, NSW, Australia Cun-Zheng Ning, Department of Electrical Engineering, Arizona State University, Tempe, AZ, USA Toyoaki Nishida, Department of Intelligence Science and Technology, Kyoto University, Kyoto, Japan Luca Oneto, Department of Informatics, Bioengineering, Robotics and Systems Engineering, University of Genova, Genova, Genova, Italy Bijaya Ketan Panigrahi, Department of Electrical Engineering, Indian Institute of Technology Delhi, New Delhi, Delhi, India Federica Pascucci, Department di Ingegneria, Università degli Studi Roma Tre, Roma, Italy Yong Qin, State Key Laboratory of Rail Traffic Control and Safety, Beijing Jiaotong University, Beijing, China Gan Woon Seng, School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore, Singapore Joachim Speidel, Institute of Telecommunications, University of Stuttgart, Stuttgart, Germany Germano Veiga, FEUP Campus, INESC Porto, Porto, Portugal Haitao Wu, Academy of Opto-electronics, Chinese Academy of Sciences, Haidian District Beijing, China Walter Zamboni, Department of Computer Engineering, Electrical Engineering and Applied Mathematics, DIEM—Università degli studi di Salerno, Fisciano, Salerno, Italy Junjie James Zhang, Charlotte, NC, USA Kay Chen Tan, Department of Computing, Hong Kong Polytechnic University, Kowloon Tong, Hong Kong

The book series Lecture Notes in Electrical Engineering (LNEE) publishes the latest developments in Electrical Engineering—quickly, informally and in high quality. While original research reported in proceedings and monographs has traditionally formed the core of LNEE, we also encourage authors to submit books devoted to supporting student education and professional training in the various fields and applications areas of electrical engineering. The series cover classical and emerging topics concerning: • • • • • • • • • • • •

Communication Engineering, Information Theory and Networks Electronics Engineering and Microelectronics Signal, Image and Speech Processing Wireless and Mobile Communication Circuits and Systems Energy Systems, Power Electronics and Electrical Machines Electro-optical Engineering Instrumentation Engineering Avionics Engineering Control Systems Internet-of-Things and Cybersecurity Biomedical Devices, MEMS and NEMS

For general information about this book series, comments or suggestions, please contact [email protected]. To submit a proposal or request further information, please contact the Publishing Editor in your country: China Jasmine Dou, Editor ([email protected]) India, Japan, Rest of Asia Swati Meherishi, Editorial Director ([email protected]) Southeast Asia, Australia, New Zealand Ramesh Nath Premnath, Editor ([email protected]) USA, Canada Michael Luby, Senior Editor ([email protected]) All other Countries Leontina Di Cecco, Senior Editor ([email protected]) ** This series is indexed by EI Compendex and Scopus databases. **

B. K. Murthy · B. V. R. Reddy · Nitasha Hasteer · Jean-Paul Van Belle Editors

Decision Intelligence Proceedings of the International Conference on Information Technology, InCITe 2023, Volume 1

Editors B. K. Murthy Innovation and Technology Foundation (IBITF) Indian Institute of Technology Bhilai Sejbahar, Chhattisgarh, India Nitasha Hasteer Amity School of Engineering and Technology Amity University Uttar Pradesh Noida, Uttar Pradesh, India

B. V. R. Reddy NIT Kurukshetra Thanesar, Haryana, India Jean-Paul Van Belle Department of Information System University of Cape Town Cape Town, South Africa

ISSN 1876-1100 ISSN 1876-1119 (electronic) Lecture Notes in Electrical Engineering ISBN 978-981-99-5996-9 ISBN 978-981-99-5997-6 (eBook) https://doi.org/10.1007/978-981-99-5997-6 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd. The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721, Singapore Paper in this product is recyclable.

Preface

Decision intelligence is a discipline that combines data science, machine learning and behavioral science to help people and organizations make better decisions. It involves the use of advanced analytics and decision-making techniques to generate insights and recommendations that are grounded in data and aligned with human values and preferences. Decision intelligence applies a systematic approach to decision-making that considers multiple options, evaluates trade-offs and incorporates feedback and learning to continuously improve decision-making over time. The ultimate goal of decision intelligence is to improve the quality and speed of decision-making while minimizing risks and uncertainties. It can be applied to a wide range of decision domains, including health care, education, governance and business. This book is a compilation of research work carried out by esteemed experts in the area of decision intelligence. The objective of this compilation is to provide relevant and timely information to those who are striving to contribute to this domain by utilizing the latest technology. The content of this book is based on the research papers accepted during the 3rd International Conference on Information Technology (InCITe 2023) held at Amity University, Uttar Pradesh, Noida, India, during March 2 and 3, 2023. It is a conglomeration of research papers covering interdisciplinary research and in-depth applications of AI and machine learning for decision-making, intelligent system design and modeling, intelligent networks and security, image processing and cognition systems, software engineering and quality management. The content would serve as a rich knowledge repository on information and communication technologies, neural networks, fuzzy systems, natural language processing, data mining and warehousing, big data analytics, cloud computing, security, social networks and intelligence, decision-making and modeling, information systems and IT architectures. We are hopeful that it will prove to be of high value to graduate students, researchers, scientists, practitioners and policymakers in the field of information technology. We thank the management of Amity University, Uttar Pradesh, who believed in us and provided the opportunity for publishing the book. We are grateful to Science and Engineering Research Board, Department of Science and Technology (DST), and Defence Research and Development Organisation (DRDO), Government of India, v

vi

Preface

for their valuable support. We are thankful to the advisory committee of InCITe 2023 for the continuous support and guidance. Special thanks to all the national and international reviewers who helped us in selecting the best of the works as different chapters of the book. We wish to acknowledge and thank all the authors and coauthors of different chapters who cooperated with us at every stage of publication and helped us to sail through this mammoth task. We owe our sincere gratitude to all our family members and friends who helped us through this journey of publishing a book. Our appreciation goes to each and every individual who supported us in this endeavor. Last but not the least, we are grateful to the editing team of Springer who provided all the guidance and support to us in the compilation of the book and also shaped it into a marketable product. Bhilai, India Kurukshetra, India Noida, India Rondebosch, South Africa

B. K. Murthy B. V. R. Reddy Nitasha Hasteer Jean-Paul Van Belle

Contents

A Radio Frequency-Based Energy Harvesting Model for IoMT Device . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Angkurita Roy, Noorafsha Tahseen, and Nabajyoti Medhi

1

Cancer Hotspot Identification and Analysis: A Scan Statistics Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sonia Kaindal, B. Venkataramana, and Jitendra Kumar

13

An Artificial Voice Box that Makes Use of Unconventional Methods of Machine Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Raman Chadha, Sanjay Singla, and Nongmeikapam Thoiba Singh

29

State-of-Art Review on Medical Image Classification Techniques . . . . . . . Abhishek Bose and Ritu Garg Prediction of Loan Approval of Customers Based on Credit Score Using Machine Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Navya Gupta, Deeksha Sharma, Alka, Vrinda Aggarwal, Veronica Singh, Poonam Bansal, and Kiran Malik Strategies for the Adoption of AI Technologies in the South African Wine and Fruit Industries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Robert Walton, Jean-Paul Van Belle, and Nitasha Hasteer Real-Time Facial Mask Detection Using Deep Learning . . . . . . . . . . . . . . . Ayush Chauhan, Rohan Rajput, Divyansh Chaudhary, and Aakanshi Gupta SSATS—Enhancement of Semantic Similarity of Abstractive Text Summarization Using Transformer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . R. Thirisha, A. Subarna Kiruthiga, S. Arunkumar, and J. Felicia Lilian Semantic Segmentation of Optical Satellite Images . . . . . . . . . . . . . . . . . . . Yashasvi Mehta, Vijay Katkar, and ShobhitKumar Patel

43

53

63 75

87 99

vii

viii

Contents

Human Gender and Age Estimator Using Local and Global Features with Machine Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 Jayaprada S. Hiremath and Shantakumar B. Patıl Predictive Analysis of Neurodegenerative and Chronic Fatigue Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123 Rishi Raj, Jagriti Pal, Kumar Kriti, and Nikita Verma An Automatic Early Alert System on Detecting Dozing Driver . . . . . . . . . 133 Indu Chawla, Archana Purwar, Shyam Agarwal, Sankalp Agrawal, and Ria Ahlawat Reversible Image Steganography to Achieve Effective PSNR . . . . . . . . . . . 145 Mohammad Gauhar Nayab, Aditya Pratap Singh, Ritik Sharma, and Gaurav Raj EDA and Predicting Customer’s Response for Cross-Sell Vehicle Insurance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157 Anand Jha, Jankisharan Pahareeya, Kirtiraj Bhatele, and Sanjay Patsariya A Study of Feature-Based and Pixel-Level Image Fusion Techniques . . . 169 Vivek Kumar, Manisha Khanduja, Harishchander Anandaram, Kapil Joshi, Ashulekha Gupta, and Manoj Diwakar Catch Your Session, Track Your Pulse e-Health Service Using Blockchain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179 N. Sharmili, K. Chandu, J. N. V. S. Harshitha, B. Aishwarya, and B. Kumari Ansuni Baat—A Bridge to Sign Language Communication . . . . . . . . . . . . 191 Jai Narain, Shivam Singh, Siddhant Srivastava, and Purushottam Sharma A Review of Deep Learning Algorithms for Early Detection of Oral Mouth Cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203 Yomesh Sharma and Jagdeep Kaur Malware Detection Using Deep Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215 Mridul Vohra, Anushka Tiwari, Purushottam Sharma, and Ayush Jayaswal Automatic Title Generation with Attention-Based LSTM . . . . . . . . . . . . . . 233 M. Dhilsath Fathima, M. Seeni Syed Raviyathu Ammal, Prashant Kumar Singh, Sachi Shome, Manbha Kharsyienlieh, and R. Hariharan COVID-19 (COV-19) Spreading Diagnoses by Feature Representation Method Through Visual Learning (FVisL-CoV19) . . . . . 243 V. Kakulapati and A. Jayanthiladevi

Contents

ix

Alzheimer’s Disease Classification Using Deep Learning Models . . . . . . . 257 N. Rajathi and G. Malavika Brain Pathology Detection Using Convolutional Neural Network from EEG Signal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265 M. Kavitha, J. Janet Cyndhiya, R. Srinivasan, and R. Kavitha CNN-Based Adversarial Embedding for Image Steganograpy . . . . . . . . . 275 Amit, Aditya Agarwal, Saptarshi Gupta, Minakshi Sanadhya, Arun Kumar, and Abhilasha Singh Machine Learning Approaches for Entity Extraction from Citation Strings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287 Vidhi Jain, Niyati Baliyan, and Shammy Kumar A Survey on Brain Tumor Segmentation with Missing MRI Modalities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299 Deep Shah, Amit Barve, Brijesh Vala, and Jay Gandhi Deep Learning Based Segmentation Approach for Cervical Cancer Detection Using Pap Smears . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309 C. V. Sagar, Harshit Bhardwaj, and Anupama Bhan Hybrid Sign Language Interpreter Development Using Machine Learning Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319 Dev Walia, Himanshu Jakhmola, Akull Nainwal, and Sanjay Kumar Dubey Community Detection Algorithms in Social Networks: An Empirical Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333 Law Kumar and Rajeev Kumar Designing Hybrid Image Fusion Algorithm Using CNN and Stationary Wavelet Transform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 345 Rishabh Sharma and Ashok Kumar Yadav Predict Pawpularity Score of Pets Using State of Art Algorithms . . . . . . . 357 Abhigya Verma, Astha Bhaskar, Pooja Gera, and Shweta Singhal Comparative Analysis of Balanced Code Smell Detection Using Machine Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 371 Meghna Sabharwal, Aakankshi Gupta, Rashmi Gandhi, and Ishan Khan ETL Data Pipeline to Analyze Scraped Data . . . . . . . . . . . . . . . . . . . . . . . . . 379 Rashmi Gandhi, Sparsh Khurana, and Harsh Manchanda DDoS Attack Detection Using Machine Learning . . . . . . . . . . . . . . . . . . . . . 389 Jadhav Swati, Pise Nitin, Shruti Singh, Akash Sinha, Vishal Sirvi, and Shreyansh Srivastava

x

Contents

An Empirical Study to Investigate Class Imbalance Issue for Improving Security Bug Report Classification Prediction . . . . . . . . . . 405 Rashmi and Arvinder Kaur Factors Influencing the Use of a Soccer Team’s Mobile Application by Fans in South Africa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 417 Athir Fayker and Jean-Paul Van Belle

About the Editors

Dr. B. K. Murthy is CEO of the Innovation and Technology Foundation at the Indian Institute of Technology (IIT) Bhilai. He was Scientist G and Group Coordinator (R&D in IT) in the Ministry of Electronics and IT (MeitY), Government of India, where he was responsible for the promotion of R&D in the area of IT. He has been conferred the prestigious VASVIK Industrial Research Award for the year 2020 in the category of Information and Communication Technology. He spearheaded bringing out the National Strategy on Blockchain Technology from the Ministry of Electronics and IT which was released in December 2021. His research interests include artificial intelligence, software-defined networking, cloud computing, quantum computing, and blockchain technologies. He was awarded his Ph.D. degree by IIT Delhi. He has published and presented more than 70 papers in various journals and conferences and is a regular speaker at international forums. Dr. B. V. R. Reddy is the Director of the National Institute of Technology (NIT) Kurukshetra. Earlier he served as Professor at the University School of Information & Communication Technology (USICT), GGSIP University, India, for 22 years. He possesses an excellent academic track record both as a teacher and as a researcher besides administrative acumen in technology areas of Information and Communication Technology (ICT). He has also served in various positions as Dean of School of Engineering and Technology (USET), Dean of University School of Information and Communication Technology (USICT), Dean of University School of Architecture and Planning (USAP), Chairman and Librarian for University Information Resource Centre (UIRC), served at national institutes of repute as Faculty at NIT Kurukshetra, NIT Hamirpur and an alumnus of Andhra University, IIT Roorkee and NIT Kurukshetra. Dr. Nitasha Hasteer is Professor & Head of the Information Technology Department and Dy. Director (Academics) at Amity School of Engineering & Technology, Amity University, India. She has 21 years of teaching, research, and administrative experience in academics and industry and is a Ph.D. in Computer Science and Engineering. Her interest areas are machine learning, cloud computing, crowdsourced xi

xii

About the Editors

software development, software project management, and process modeling through multi-criteria decision-making techniques. She has published more than 65 papers in various journals and international conferences in these areas and has guided many postgraduate students in their dissertations. She has been on the editorial board of many international conferences and obtained funding from government agencies such as the Science and Engineering Research Board, Department of Science Technology (DST), Defence Research & Development Organization (DRDO), Indian National Science Academy (INSA), and Council of Scientific & Industrial Research (CSIR) for organizing conferences in the area of Information Technology. She is a member of the International Association of Computer Science & Information Technology (IACSIT), Singapore, the Institution of Engineering and Technology (IET), UK, and various other professional bodies. Dr. Jean-Paul Van Belle is Professor in the Department of Information Systems, University of Cape Town, South Africa and has been the Director of the Centre for IT and National Development in Africa (CITANDA). His particular research focus and passion is the adoption and appropriation of emerging ICTs in a development context i.e. Development Informatics, ICT4D and Mobile for Development (M4D). His other research and teaching specializations are Social Networking, Decision Support, Business Analytics, Open Government Data, E- & M-commerce, E- & Mgovernment, Organizational Impacts and Adoption of IS, Open-Source Software, IT/IS Architectures and Artificial Intelligence. He has active research collaborations with researchers in India, UK, Ethiopia, Kenya, Ecuador and Chile.

A Radio Frequency-Based Energy Harvesting Model for IoMT Device Angkurita Roy , Noorafsha Tahseen , and Nabajyoti Medhi

Abstract The Internet of Medical Things (IoMT), also known as Smart Healthcare, has seen incredible progress in the Smart Environment industry. A significant part of Industry 4.0 is Healthcare 4.0, which is transforming the medical industry to monitor patients’ health remotely and perform other health-related activities. The IoMT includes many wearable sensors that either work on batteries or are rechargeable. In this paper, we work to harvest ambient sources of energy and utilize them for powering medical sensor batteries. Energy Harvesting is one such technique of capturing energy from the surroundings (ambient energy or non-ambient energy) and then converting it to usable electrical energy. We designed a model for Energy Harvesting and IoMT where we take radio frequency as an ambient energy source for harvesting to power the sensors. As medical sensors are mostly used indoors, they are more exposed to radio frequency, which is an ideal source of energy. Finally, simulation results have been obtained showing that the antenna can absorb radio frequency energy up to 50 ohms depending on the impedance of the circuit. Also, the rectifier could convert the AC signal to a DC signal which could be boosted by a voltage regulator to 47 V meeting the requirement to charge a lithium-ion battery. Keywords IoMT · IoT (Internet of Things) · Radio Frequency Energy Harvesting (RFEH)

A. Roy (B) · N. Medhi Department of Computer Science and Engineering, Tezpur University, Tezpur, Assam, India e-mail: [email protected] N. Medhi e-mail: [email protected] N. Tahseen Department of Electrical Engineering, Tezpur University, Tezpur, Assam, India e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 B. K. Murthy et al. (eds.), Decision Intelligence, Lecture Notes in Electrical Engineering 1079, https://doi.org/10.1007/978-981-99-5997-6_1

1

2

A. Roy et al.

1 Introduction Energy comes from the Greek word “Energeia” and is a quantitative property that is found in any state. According to the Law of Conservation, energy cannot be created or destroyed. It can only be transformed from one form of energy to another form. The place where we live is filled with a variety of energies, such as solar energy, body heat energy, and human movement energy. All the aforementioned energies can be found from various sources which can be captured and utilized for various purposes [7, 10]. Energy Harvesting, a technique of Green-IoT, is a process that involves capturing the unexploited energies in our environment and transforming them into electrical energy to use further [20, 28]. A Green-IoT solution proposes to resolve energy utilization problems of IoT to create a sustainable and green environment [3, 22]. One of the leading technologies, IoT, has been incorporated into each field to create a smarter environment. The application of IoT can be seen in every field such as industry, home, health care, transportation, and agriculture [1]. IoT is basically the interconnection of many devices using different applications. Every node in IoT devices operates on a chargeable basis. Sensors in IoT require a significant amount of energy, being the most critical part of IoT [6]. Similarly in IoMT, medical sensors consume a lot of energy [29]. Energy Harvesting, therefore, enters into the picture, a technique that enables low-powered devices to be powered by ambient sources of energy, reducing pollution and moving to Green-IoT [14, 18]. IoT in health care is revolutionizing the healthcare sector, considering Healthcare 4.0 challenges and benefits [17]. The charging factor is a key factor in making IoT devices more reliable. IoT devices with rechargeable batteries or non-rechargeable batteries require a certain amount of energy to keep them working. Efficient power consumption has always been a concern. Software-Defined based fog network for quality of service provisioning of data collection from healthcare sensors which form the end-device layer does consume a significant amount of energy [2, 12]. Sensors for data collection require continuous power. This system uses ambient energy, i.e. radio frequency (RF) energy that is preferable in closed rooms, to harvest it for charging various IoMT devices. The RF energy may come from reliable sources (RFID chips), ambient sources (radio, Wi-Fi access points), or RF energy transfer between mobile devices. Figure 1 shows the various RF energy sources converted to electrical energy to be used by IoMT devices.

Fig. 1 RF energy sources converting to electrical energy for medical sensors [8]

A Radio Frequency-Based Energy Harvesting Model for IoMT Device

3

1.1 Contribution The contribution in the paper includes the following: The EH model is an Energy Harvesting model that can absorb RF energy from various RF sources. The model consists of an antenna, an impedance network, a rectifier, a voltage regulator, and a rechargeable battery. The IoMT model is an IoMT-based smart healthcare system that consists of IoMT devices primarily medical sensors that use harvested energy to function. Also, a comprehensive analysis of sensor power consumption data will be extracted and optimized using algorithms and machine learning techniques. In the rest of the paper, Section 2 describes related works on Radio Frequency harvesting. Section 3 describes the proposed model. Section 4 evaluates the model, and Section 5 concludes the work.

2 Related Works This section discusses works related to Energy Harvesting that have been carried out in recent years. IoT is connecting everything from anywhere at any time. IoMT, a subset of IoT, should be pushed out and revolutionized by incorporating the rules of Industry 4.0 resolving the challenges of Healthcare 4.0 for a Smart Healthcare system. Rehman et al. [17] discussed the drift, challenges, and gaps of Healthcare 4.0, i.e. incorporating Industry 4.0 into the healthcare industry. It is a concept with challenges in terms of security and privacy, feasibility, scalability, interoperability, and network limitations. The study mainly focused on bringing forward solutions which could help in acquiring Industry 4.0 technologies. Pozo et al. [15] studied the various ambient energies present in and around the environment which discussed different harvesters or models required to capture those energies. Among them are Photovoltaic, Piezoelectric, Electromagnetic, Electrostatic, RF, Wind, Water, Acoustic, Magnetic, and various other energies, which constitute a feasible solution to Green-IoT. Zeadally et al. [31] reported major design architectures for harvesting energy along with storage devices and control units. In addition to 6 major energy resources, it includes energy harvesters that can be deployed in IoT environments. Also, it discusses various challenges related to the design architecture of the energy harvester for its smooth functioning. Filios et al. [26] presented a platform where Energy Harvesting techniques, energy storage, and energy management for both rechargeable and non-rechargeable batteries were presented. It embeds an efficient Energy Harvesting circuit for low-power and miniature energy generators. Sanislav et al. [19] discussed Energy Harvesting technology, which can often extend the battery life of low-powered devices or replace the entire battery. They

4

A. Roy et al.

examined a few cases of harvesting techniques and suggested deploying such techniques on a large scale. Wang et al. [30] made an overview of recent research techniques and challenges in various energy management solutions. In addition, IoT power consumption techniques were studied and various machine learning, deep learning, and clustering techniques for energy management solutions were evaluated. Theilakarathne et al. [25] explained Green-IoT, a concept of creating a sustainable environment. IoT is a network of millions of devices, but it consumes a lot of energy and wastes it as heat. In order to overcome the aforementioned disadvantage, the concept of Green-IoT has been introduced in IoT devices for easier applicability. Liu et al. [9] studied the evolution of Energy Harvesting from fundamental to material. The authors studied current strategies for Energy Harvesting on various self-powered devices and how they can be applied to various fields, such as human monitoring and robotic transportation. Sherazi et al. [23] explained the development of Energy Harvesting fields and their application in various environments, where sensor networks are deployed. The authors report the limited number of studies in this particular area, and the challenges it poses to RF Energy Harvesting.

3 Proposed Model In the proposed model, we aim to develop a reliable, sustainable, and efficient radio frequency Energy Harvesting system for IoMT sensors that can work on a regular basis to monitor a patient’s vital signs in a hospital, keeping a record of the patient and sending data to the cloud for further analysis contributing to Green-IoT. The model consists of 2 parts: The Radio Frequency Energy Harvesting (RFEH) model and the Radio Frequency Energy Harvesting-Internet of Medical Things (RFEH-IoMT) model.

3.1 RFEH Model The RFEH model is basically an environmental or ambient Energy Harvesting model. It absorbs radio frequency energy from various RF sources found around the environment and converts it to electrical energy for its purposes. The proposed model is different from other existing models as it converts ambient radio frequency energy to electric current [8]. The RFEH model mainly comprises a receiving antenna capturing the RF signal from nearby sources. Impedance-matching network plays a vital role in transferring maximum power between the source and destination. A rectifier converts the RF energy harvested by the antenna and the received RF signal into direct current (DC). A Voltage Regulator is used for maintaining fixed output voltage and storing

A Radio Frequency-Based Energy Harvesting Model for IoMT Device

5

Fig. 2 Microstrip patch antenna

Fig. 3 Antenna pattern graph

the power in a capacitor or battery for further usage by IoMT sensors [16]. Figure 4a displays the RFEH-Circuit Model in dotted lines. Antenna The RF energy is absorbed or received by an antenna which can be of any type such as a dipole antenna, patch antenna, and rectangular patch antenna. Here, in our MATLAB simulation environment shown in Fig. 2. RFEH-circuit, we have chosen microstrip patch antennas as they can be easily fabricated with printed circuit board techniques and can be made easily to most surface profiles [5, 13]. Figure 3 shows the Antenna Pattern Graph of the microstrip patch antenna. They can transmit/ receive in any polarization [21]. The antenna is designed with impedance matching to support 50 ohms. An antenna is selected based on its size, material, polarization, frequency, sensitivity, and directivity. The polarization and frequency of ambient sources cannot be predicted but can be predicted for non-ambient sources. The size of an antenna is mainly justified on the basis of its compact and portable nature as well as its good frequency response. While sensitivity depicts the harvesting of energy and operating at low input power. The directivity feature is an omnidirectional radiation pattern for the ambient source, while the unidirectional radiation pattern is for the non-ambient source. For any antenna, the features that the design of the antenna depends on are shown in Table 1 [27]. Impedance Matching The impedance-matching network receives the signals from the antenna and transfers them to the rectifier circuit. An impedance circuit is crucial for bringing the antenna and rectifier on the same pitch for the smooth operation of the system. The impedance is the addition of resistance and reactance from the receiving antenna. When the impedance of the rectifier does not match the impedance of the antenna, then a proportion of energy is not absorbed from the source and reflected

6

A. Roy et al.

(a)

(b)

Fig. 4 RFEH-IoMT model and proposed methodology

Table 1 Overview of types of antennas Type of antenna

Frequency

Size

Sensitivity

Directivity

mm2

Low with −16 dB gain

Depends on ambient or non-ambient

Microstrip patch antenna

2.4 GHz

38

Patch antenna

2.45 GHz

20*20 mm2

Low with −25 dB gain but less efficient

Depends on ambient or non-ambient

Microstrip-fed dipole

900 MHz

62*62 mm2

Low with 1.84 dB gain

Depends on ambient or non-ambient

Rectangular patch

2.45 GHz

100*100*5 mm3

Low with 8.35 dB gain

Depends on ambient or non-ambient

CPW fed patch

866 MHz

101.8*46.5 mm2

Low with 2 dB gain

Depends on ambient or non-ambient

Folded dipole

918 MHz

30*30*10 mm3

Low with 1 dB gain

Depends on ambient or non-ambient

back. The complex source impedance, i.e. RSource + jXSource should match load impedance, i.e. RLoad + j0 which is achieved with a matching network (-jXMatch ) with equal and opposite reactance from the source. Thus, giving the result of the impedance network as [4]: R Source= R Load If the impedance of the source and the impedance of the rectifier match, then we will have zero percent loss of energy. Rectifier The working principle of the rectifier in our model is to convert RF energy to DC voltage, only when it can be stored in a battery. Diodes are used in the rectifier system because they increase the overall efficiency of the circuit and prevent energy loss.

A Radio Frequency-Based Energy Harvesting Model for IoMT Device

7

Power conversion efficiency (PCE) is the RF to DC conversion which is defined as the amount of energy transferred from the device [4]: n=P

in/P out

Pout = V2out /RL where Pin is RF energy (input power), Pout (output power) with Vout (output voltage), and RL (output load). Voltage Regulator A voltage regulator is used for the constant and required amount of power supply to the energy storage holder. It is used to improve the capability of providing voltage in RF Energy Harvesting systems as required and have lower drop voltage [4]. Battery Storage Battery storage or energy storage is the most significant requirement of the model since it stores energy. It has the capacity to convert electrical energy to chemical energy and vice versa as and when required. Once the energy is loaded into the battery, it can be used by any sensor [21], whenever required.

3.2 RFEH-IoMT Model In the RFEH-IoMT model, there is a sensor node and an edge node that communicate with the cloud server. IoMT is one of the applications of Energy Harvesting required for the IoMT device [24]. Sensor power consumption data is stored in the cloud primarily for exploratory analysis of data. A comparison of the dataset with normal power consumption will be made. For any differences, optimization techniques using machine learning techniques will be focused on, in order to create an efficient Energy Harvesting IoMT model. Figure 4a shows the IoMT model combined with the RFEH model. Based on the proposed methodology shown in Fig. 4b, we aim to create an RF energy harvester shown in Fig. 4a. After harvesting RF energy in the RFEH model, the IoMT device will function based on the energy stored or harvested. As a result, the IoMT-based smart healthcare model will create an energy-efficient network and increase the performance of the IoMT device in a diverse environment. The energy consumed by sensors will be examined and the application of various techniques will be carried out for optimization.

8

A. Roy et al.

4 Performance Analysis In this section, we have analyzed our work in a MATLAB environment based on circuit diagrams storing the required energy in a rechargeable battery. A microstrip antenna has a substantial absorption power of radio frequency energy, and an impedance of 50 ohms at maximum meets the impedance of the complete RFEHcircuit. If the impedance is lesser than 50 ohm, then it would not meet the impedance of the circuit giving low voltage energy. The Antenna Impedance Graph and Sparameters Graph depict the impedance and effective absorption of RF energy. The S-parameters S11 and S21 represent the most efficient radiation of RF energy and absorption of energy, resulting in a fully charged battery for turning on a sensor. Antenna Impedance Graph Antenna impedance is crucial and imperative for effective communication. To achieve maximum power transfer from the antenna to the receiver or transmitter to the antenna, it should match the system impedance as well [11]. Antenna impedance includes both reactance and resistance. Resistance is the amount of energy that is absorbed from the near field of the antenna. Reactance is the amount of power stored in the near field of the antenna. Figure 5a shows that the impedance result was near 50 ohms at 2.4 GHz, which is in accordance with the impedance network. S-parameters Graph The S-parameter graph illustrates the input-output relationship between ports in an electrical system. They speak about the propagation of energy through an electrical network. In Fig. 5b, the graph has two coordinates: magnitude and frequency. S11 represents how much power is reflected from the antenna. At 2.4 GHz, the antenna radiates best when S11 = −16dB and it is least when it reaches −2 dB. Cartesian Graph of Impedance Matching An impedance-matching network connects an antenna to a rectifier circuit, allowing the source and load of the same amount of energy. Figure 5c illustrates the Cartesian graph of the impedancematching network showing two parameters: S11 shows reflected power from the antenna, which is a reflection coefficient, and S21 is a transmission coefficient. According to the graph, the amount of energy absorbed has been transmitted without loss. Rectifier A rectifier is a device that converts an alternating current (AC) to a direct current (DC). In Fig. 6a and b, we see that an AC signal varying between −100 and 100v is rectified to a DC signal varying between 0 and 100.

(a)

(b)

Fig. 5 Impedance graph, S-parameters graph, and Cartesian graph

(c)

A Radio Frequency-Based Energy Harvesting Model for IoMT Device

(a)

(b)

(c)

9

(d)

Fig. 6 Input/output of rectifier and voltage regulator

Graph of Voltage Regulator The voltage regulator improves the input DC voltage to a specified DC output voltage. Figure 6c and d show scope 1 of the voltage regulator, a graph with time in the x-axis of given 10 seconds and voltage in the y-axis of 12 V which has been boosted to about 47 V. The antenna impedance graph shows that the resistance at 2.4 GHz has the highest energy absorption, which is greater than 50 ohms. This is because it meets the impedance of the circuit, which transfers the energy to the rectifier with the same impedance. The S-parameters graph indicates that the antenna radiates best at 2.4 GHz. Using the impedance-matching graph, we can determine that the circuit can pass the required energy. AC current is converted to DC by the rectifier that matches the same impedance as the antenna. The voltage regulator improves the DC necessary to charge the sensor battery. This results in the amount of DC needed to charge a battery. IoMT sensors will be powered by the RFEH model, which will meet Green-IoT goals.

5 Conclusion In this paper, we focused on the simulation of the radio frequency-based Energy Harvesting model and its implementation. Radio Frequency, an ambient energy source chosen as an alternative to electricity for Energy Harvesting, will contribute to Green-IoT. The simulation described above was carried out entirely in a MATLAB environment. The future research will be wholly implemented on a hardware-based setup, and ultimately the entire process of Energy Harvesting will be tested for IoMT healthcare sensors. Additionally, power optimization techniques and efficiency on sensors will be investigated experimentally. Moreover, exploratory analyses of the sensor’s power consumption data will also be interpreted in the future. Acknowledgements This work is financially supported by the Ministry of Electronics and Information Technology (MeitY), India, with project no. 13(15)/2020-CC&BT dated 21.07.2020.

10

A. Roy et al.

References 1. Bajaj A, Bhatnagar M, Chauhan A (2021) Recent trends in internet of medical things: a review. Adv Machine Learn Comput Intell 645–656 2. Bardalai P, Medhi N, Bargayary B, Saikia DK (2021) Openhealthq: Openflow based QOS management of healthcare data in a software-defined fog environment. In: ICC 2021—IEEE international conference on communications, pp 1–6. https://doi.org/10.1109/ICC42927.2021. 9500637 3. Benhamaid S, Bouabdallah A, Lakhlef H (2022) Recent advances in energy management for green-IoT: an up-to-date and comprehensive survey. J Netw Comput Appl 198:103257 4. Bougas ID, Papadopoulou MS, Boursianis AD, Kokkinidis K, Goudos SK (2021) State-of-theart techniques in rf energy harvesting circuits. In: Telecom, vol 2, pp 369–389. MDPI 5. El Alami A, Bennani SD, El Bekkali M, Benbassou A (2013) Optimization and high gain of a microstrip patch antenna excited by coaxial probe for RFID reader applications at 2.4 Ghz. Eur J Sci Res 104(3):377–391 6. Garg N, Garg R (2017) Energy harvesting in IoT devices: a survey. In: 2017 international conference on intelligent sustainable systems (ICISS). IEEE, pp 127–131 7. Haseli Y (2020) Chapter one - fundamental concepts. In: Haseli Y (ed) Entropy analysis in thermal engineering systems. Academic Press, pp 1–11. https://doi.org/10.1016/ B978-0-12-819168-2.00001-5. https://www.sciencedirect.com/science/article/pii/B97801281 91682000015 8. Ibrahim HH, Singh MJ, Al-Bawri SS, Ibrahim SK, Islam MT, Alzamil A, Islam MS (2022) Radio frequency energy harvesting technologies: a comprehensive review on designing, methodologies, and potential applications. Sensors 22(11):4144 9. Liu L, Guo X, Liu W, Lee C (2021) Recent progress in the energy harvesting technology—from self-powered sensors to self-sustained IoT, and new applications. Nano-materials 11(11):2975 10. Muhammad S, Tiang JJ, Wong SK, Rambe AH, Adam I, Smida A, Waly MI, Iqbal A, Abubakar AS, Mohd Yasin MN (2022) Harvesting systems for rf energy: trends, challenges, techniques, and tradeoffs. Electronics 11(6):959 11. Nikolova NK, Ravan M, Amineh RK (2012) Substrate-integrated antennas on silicon. In: Advances in imaging and electron physics, vol 174, pp 391–458. Elsevier 12. Osifeko MO, Hancke GP, Abu-Mahfouz AM (2020) Artificial intelligence techniques for cognitive sensing in future IoT: state-of-the-art, potentials, and challenges. J Sensor Actuator Netw 9(2):21 13. Ossa-Molina O, López-Giraldo F (2022) A simple model to compute the characteristic parameters of a slotted rectangular microstrip patch antenna. Electronics 11(1):129 14. Panahi FH, Moshirvaziri S, Mihemmedi Y, Panahi FH, Ohtsuki T (2018) Smart energy harvesting for internet of things. In: 2018 smart grid conference (SGC), pp 1–5. https://doi. org/10.1109/SGC.2018.8777889 15. Pozo B, Garate JI, Araujo JÁ, Ferreiro S (2019) Energy harvesting technologies and equivalent electronic structural models. Electronics 8(5):486 16. Rajput S, Averbukh M, Rodriguez N (2022) Energy harvesting and energy storage systems 17. Rehman MU, Andargoli AE, Pousti H (2019) Healthcare 4.0: trends, challenges and benefits 18. Richards J, Lim M, Li G, Araya E, Jia Y (2020) Continuous ECG monitoring with low-power electronics and energy harvesting. In: 2020 IEEE 63rd international midwest symposium on circuits and systems (MWSCAS), pp 643–646. https://doi.org/10.1109/MWSCAS48704.2020. 9184610 19. Sanislav T, Mois GD, Zeadally S, Folea SC (2021) Energy harvesting techniques for internet of things (IoT). IEEE Access 9:39530–39549 20. Sarkar S, Debnath A (2021) Green IoT: design goals, challenges and energy solutions. In: 2021 6th international conference on communication and electronics systems (ICCES). IEEE, pp 637–642

A Radio Frequency-Based Energy Harvesting Model for IoMT Device

11

21. Sharma DK, et al (2022) Comparative analysis of rectenna systems for radio frequency energy harvesting. In: New trends and applications in Internet of Things (IoT) and big data analytics. Springer, pp 61–83 22. Sharma N, Panwar D (2020) Green IoT: advancements and sustainability with environment by 2050. In: 2020 8th international conference on reliability, infocom technologies and optimization (Trends and Future Directions) (ICRITO). IEEE, pp 1127–1132 23. Sherazi HHR, Zorbas D, O’Flynn B (2022) A comprehensive survey on rf energy harvesting: applications and performance determinants. Sensors 22(8):2990 24. Srivastava J, Routray S, Ahmad S, Waris MM (2022) Internet of medical things (IOMT)-based smart healthcare system: trends and progress. Comput Intell Neurosci 2022 25. Thilakarathne NN, Kagita MK, Priyashan W, Green Internet of Things: the next generation energy efficient internet of things. In: Applied information processing systems. Springer, pp 391–402 26. Tsenempis I, Filios G, Katsidimas I, Nikoletseas S (2019) Energy harvesting and smart management platform for low power IoT systems. In: 2019 15th international conference on distributed computing in sensor systems (DCOSS). IEEE, pp 339–346 27. Ullah MA, Keshavarz R, Abolhasan M, Lipman J, Esselle KP, Shariati N (2022) A review on antenna technologies for ambient rf energy harvesting and wireless power transfer: designs, challenges and applications. IEEE Access 28. Verma G, Prakash S (2020) A comparative study based on different energy saving mechanisms based on green internet of things (GIoT). In: 2020 8th international conference on reliability, infocom technologies and optimization (Trends and Future Directions) (ICRITO). IEEE, pp 659–666 29. Wahba MA, Ashour AS, Ghannam R (2020) Prediction of harvestable energy for self-powered wearable healthcare devices: filling a gap. IEEE Access 8:170336–170354 30. Wang D, Zhong D, Souri A (2021) Energy management solutions in the internet of things applications: technical analysis and new research directions. Cogn Syst Res 67:33–49 31. Zeadally S, Shaikh FK, Talpur A, Sheng QZ (2020) Design architectures for energy harvesting in the Internet of Things. Renew Sustain Energy Rev 128:109901

Cancer Hotspot Identification and Analysis: A Scan Statistics Approach Sonia Kaindal , B. Venkataramana , and Jitendra Kumar

Abstract Nearly 10 million deaths from cancer are expected worldwide by 2020, making it the most common cause of death. India is anticipated to have 2.7 million cases. The count of 13.9 lakh new cases and 8.5 lakh deaths each year is highly disappointing. We proposed the application of scan statistics aiming at the identification of the hotspots of cancer incidence and mortality. The concept of hotspots allows us to develop an effective plan at the local and regional levels to combat the problems that are resulting from the concerning rise in cancer incidence and mortality. It is important to utilize spatio-temporal and longitudinal data to characterize the problem and evolve the strategy. The treatment of cancer is costly, and the consequential mental agony of the patient and their family is a great challenge to our society. The lack of health infrastructure for cancer treatment and non-optimal utilization of inadequate resources pose a great challenge and need to be resolved urgently. To address this issue, we found the hotspots of cancer incidence and mortality. We used SaTScan, MS Solver, R, SPSS, SAS, Tableau, and MS Excel to achieve computational efficacy. It is interesting to highlight that Kerala is the incidence hotspot, but Bihar is the mortality hotspot in India. The most frequently occurring cancer in males and females are the esophagus and larynx, respectively. The current work is relevant to individual patients as well as the concerned governments, policymakers, medical professionals, and all other stakeholders. Keywords Scan-statistics · Spatio-temporal · Cancer incidence and mortality · Cancer sites

S. Kaindal (B) · B. Venkataramana · J. Kumar Department of Mathematics, School of Advanced Sciences, Vellore Institute of Technology, Vellore 632014, India B. Venkataramana e-mail: [email protected] J. Kumar e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 B. K. Murthy et al. (eds.), Decision Intelligence, Lecture Notes in Electrical Engineering 1079, https://doi.org/10.1007/978-981-99-5997-6_2

13

14

S. Kaindal et al.

Abbreviations SCI UIAI RCRS APC EAPC PSA RDTs PSO

State Cancer Institute Unique Identification Authority of India Report Civil Registration System by Government Age-Period-Cohort Estimated Annual Percent Change Prostate-Specific Antigen Rapid Diagnostic Test Particle Swarm Optimization

1 Introduction In India, cancer ranks as the second most common cause of death after cardio-vascular disease. We observed relatively low awareness of cancer, a late detection rate, and unequal access to inexpensive curative services. Hippocrates (the father of medicine) thought that the disease will emerge from any inequalities in four fluids, black bile, phlegm, yellow bile, and blood, and it was believed that an abundance of black bile at a specific organ site brought in cancer, [1]. As per the study by researchers, cancer was found to be a common cause of death in middle-aged people and elderly Indians between 1917 and 1932 [2]. It is interesting to highlight the fact that from the year 1964 to 2012, the number of cancer patients in Mumbai increased by four times, according to trends from the city’s population-based cancer registry [3]. The paper is organized into five sections: Section 1 consists of an introduction in which we have denoted the research problem, related work, relevance, and objective. Section 2 presents the data design and processing in which we emphasize relevance of the secondary data sources from various government portals, research designs, and the geographical units chosen for the present study. Section 3 consists of the methodology adopted in which we have presented the methods of hotspot analysis, Welch’s T-test, consistency ranking, and Spearman Rank Correlation methods. In Sect. 4, Result, we attempted to discuss the result and related implications and tried to map these results with one objective set forth. In Sect. 5, we have presented the conclusion in the form of a summarization of our findings and an important discussion point. Figure 1 presents the flow diagram of these details.

1.1 Research Problem Cancer is becoming a serious threat to the nation, posing a significant human and economic barrier to our societal development and growth. Medical facilities, specialized hospitals, intra- and inter-departmental research centers, physicians, wards, and

Cancer Hotspot Identification and Analysis: A Scan Statistics Approach

15

Fig. 1 Flowchart

other resources connected to cancer therapy must all be optimized if cancer is to be cured. This is a key motivation and potent driving force for the present work. We aim to identify and analyze hotspot cancer sites so that minimal resources can be distributed effectively, leading to the greatest success in terms of a reduction in cancer mortality and incidence in India. The relevance of the current work is to achieve success in policy implementation, resource optimization, and optimal distribution of cancer treatment across the nation (Figs. 2 and 3).

Fig. 2 The box plot representing most consistent cancer sites among males

16

S. Kaindal et al.

Fig. 3 The box plot representing most consistent cancer sites among females

1.2 Related Work According to a study that looked at the incidence of 28 different cancers in every state from 1990 to 2016, Kerala and Mizoram had the highest crude cancer incidence rates, [4]. From 1985 to 2014, researchers when computing the cohort effect rate ratio, discovered that breast cancer rates across all Population Based Cancer Registries (PBCRs) had significantly increased. Further, a study done by Das et al. [5] provides a general picture of childhood cancer incidence in India from 2012 to 2014. Similarly, Kulothungan Estimated Annual Percent Change (EAPC) [6] analyze the cancer burden for the year 2016 using the national cancer registry programme’s forecasts for the year 2021 and 2025 (Similarly, using projections from the national cancer registry programme for 2021 and 2025, Kulothungan Estimated Annual Percent Change (EAPC) analyses the cancer burden for 2016). The AMI ratio was calculated using longitudinal NCRP-PBCR data using the Monte Carlo method (2001–2016). The results showed that the estimated number of cancer cases in India for 2021 was 26.7 million, and it was predicted that the number would increase to 29.8 million by 2025. In a study by Mohebbi et al. [7], using cancer registry data, the researchers describe the spatial geographic distribution of gastrointestinal cancer incidence and ascertain whether there were any statistically significant geographic clusters present in Iran’s Caspian region between 2001 and 2005. They discovered that esophagus and stomach tumors had non-random geographical patterns between the sexes. OrozcoAcosta et al. [8] also employed a similar approach. The paper presented by Shafer et al. [9] investigates the accuracy of several configurations of a single data set, the

Cancer Hotspot Identification and Analysis: A Scan Statistics Approach

17

county-level mortality from bone cancer in Pennsylvania from the year 1960 to 1976. Their findings showed that the lists of high-risk regions for each of the two summary indices differ significantly, directly adjusting death rate and years of life lost (r = 0.8). Some studies like [10–12] analyzed the county-level patterns of deaths in the USA. In the study by Mather et al. [13], PSA testing was implemented at the county level from 1988 to 1999, and SIRs in both racial groups were consistent with this. A study in India by Batra et al. [14] investigated the clinical presentation, investigative results, therapy, surgical findings, and outcome of 634 Gall Bladder Cancer (GBC) patient records during a ten-year period; they noted that the majority of GBC patients in India suffer from a severe, incurable disease. Jain [15] studied statistics from national population-based cancer registries on the incidence rates, mortality, and trends in prostate cancer. Bray et al. [16] have used the Human Development Index to present global cancer transitions (2008–2030) to show the patterns specialized to cancer and presented the trends from 1988 to 2002. They attempted to generate a future burden scenario for the year 2030. A study by Dikshit et al. [17] attempted to highlight tobacco-related cervical cancer deaths from the year 2001 to 2003 in rural areas of India. They find underdeveloped cancer services prevail in rural area. A study by Kamau et al. [18] presented temporal and spatial dynamics of cases of malaria with fever-positive rapid diagnostic tests. Izakian et al. [19] newly presented a scanning window optimized for identifying illness clusters using particle swarm optimization (PSO). In the study by Udbhav Bhatia et al. [20], Drowsiness Image Detection Using Computer Vision was presented, and the position of the driver’s eyes was extracted using a facial detection algorithm and facial landmark points. As a result, we may attempt to use the facial detection technique to aid in the early detection and diagnosis of cancer cases. Further, Chen et al. [21] analyzed Ontario’s incidence data geographically from 1998 to 2002; computed the p values (p = 0.01) for the Age Standard Incidence Ratio (ASIRs) among thirty-five different regions suffering from liver cancer and highlighted the city of Toronto and York as statistically significant locations. (Chen et al. also performed a geographic analysis of Ontario’s incidence data from 1998 to 2002; they calculated the p values (p = 0.01) for the Age Standard Incidence Ratios (ASIRs) among 35 different regions affected by liver cancer and highlighted Toronto and York as statistically significant locations).

2 Relevance The relevance of the present work is explained in achieving the success of policy implementation, optimal distribution of cancer treatment, and diagnostic resources to benefit the maximum number of people suffering from cancer at prior and posterior stages. We also aimed to help our government and many other stakeholders (public and private organizations) working in the direction of cancer eradication and related activities in India.

18

S. Kaindal et al.

3 Objective We have framed the following objectives: – To investigate and analyze cancer mortality, incidence, and specific sites of cancer among males and females in India by examining spatio-temporal data. – Identification of cancer incidences and mortality hotspots for the time period 2010–2020 and 2010–2018, respectively. – To test the hypothesis and see if there is a significant difference in cancer sites between males and females. – To identify the top five most consistent cancer sites for males and females during the years 2008, 2009, 2010, 2015, and 2020. – To estimate the relationship between sites of cancer among males and females.

4 Data Design and Processing Data source: secondary data from various government portals and published reports. Our research design: exploratory, descriptive, and inferential. Geographical research unit: India. In addition, we retrieved the required data set from indiastat.com. Data on cancer sites in both men and women are primarily obtained from the Ministry of Health and Family Welfare. We have used the report published from the census Total Projected Population Aadhaar Saturation (Overall) by State/UT (Unique Identification Authority of India, Government of India). In order to preserve data consistency and authenticity, we rely upon secondary data sources and avoid conducting the primary field survey.

5 Methodology Adopted 5.1 Hotspot Analysis We have used the SaTScan Kulldorff et al. [22] software to perform the computations and obtain the results, which calculated the log-likelihood ratio value for each circle and obtained the p-values through a Monte Carlo simulation process. In the present work, we used the discrete Poisson model to investigate and identify the cancer incidence and mortality hotspots spread over twenty-eight states and eight union territories (UTs) in India. Let x represent the collection of zones contained within each circle. Then scan statistics is defined mathematically as L(x) =

{(

cx μx

)Cx (

C−Cx C−μx

)C−Cx

; if cx > μx 1; otherwise

(1)

Cancer Hotspot Identification and Analysis: A Scan Statistics Approach

19

Where C total cases of death due to cancer cx total death cases in zone x that occurred in zone y μx expected cases in zone x.

5.2 Welch’s T-test It is a statistical modification of the student’s t-test, which is applied to two samples with unequal variances. The algorithm adopted is as follows: Step-1: H 0 : There is no statistically significant difference in cancer sites between males and females. H 1 : There is a statistically significant difference in sites of cancer among males and females. Step-2: Let α = 0.05, where alpha denotes the level of significance. Step-3: Computing Welch’s t-statistics: x1 − x2 tω = / 2 2 S1 S + n22 n1

(2)

where x i s , S i 2 , and ni are defined as the sample mean, variance, and size for ith data points; i=1, 2. S1 2 S2 2 + n1 n2 t(dfω ) = ( ( 2 )2 2 )2 S2 S1 n1 n2 + n1 − 1 n2 − 1

(3)

Step-4: Decision criteria: Galcern et al. [23], if the calculated t w value is greater than the critical t-values at the specified degree of freedom and level of significance, we reject H 0 ; else, we do not reject H 0 . Step-5: Making the right decision, drawing inferences, and drafting a report.

6 Spearman’s Rank Correlation Coefficient (rs ) Spearman’s rank correlation coefficient (r s ) is a correlation measure between two ranked (ordered) variables. The strength and direction of the association between two sets of data are measured using this method, Daniel et al. [24]. We set the following null and alternate hypotheses:

20

S. Kaindal et al. H 0 : There is no statistically significant relationship between the two ranks of cancer sites among males and females (defined as X and Y respectively). H 1 : There is a statistically significant relationship between the two ranks.

Spearman rank correlation coefficient value is computed as ∑ 6 di 2 rs = 1 − ; i = 1, 2, ....n n(n2 − 1)

(4)

where r s denote the Spearman rank correlation coefficient between x and y.

7 Results and Discussion The descriptive statistics is summarized in Table 1 which is used to characterize the attributes of variables under the study for both genders. The maximum value among the males is 51194, whereas it is 614404 among the females. The difference is significantly larger. We found that the mean standard error value among males is 3001.663, which is substantially lower than that of the female cancer cases (28918.630). This indicates a greater gender difference in occurrences of cancer cases. The distribution of the cases in both genders exhibits a skewness of 1.154 for men and 4.160 for females, respectively. This implies the probability distribution of cancer cases in both genders is non-symmetric. It is important to highlight that the value of kurtosis for males and females is 0.883 and 18.060, respectively. That implies a great difference in the shape and size of the distribution curve of cancer cases among the two genders. Further, the shape of the distribution curve is much more biased toward females than that for males. This refers to the alarming rising situation of cancer among females as compared to males. This needs to be controlled to achieve gender-wise success in handling cancer cases. Table 1 Descriptive statistics for sites of cancer in males and females for the year 2020 Variables

Male

Female

Variables

Male

Female

Range

41,166

609,608

Variance

144,159,644.65

17,562,029,883

Minimum

10,028

4768

Skewness

1.154

4.160

Maximum

51,194

614,404

Skewness Std. error

0.564

0.501

Mean

23,639.88

54,334.76

Kurtosis

0.883

18.060

Mean Std. error Std. deviation

3001.663 12,006.650

28,918.630 132,521.809

Skewness Std. error

1.091

0.972

Cancer Hotspot Identification and Analysis: A Scan Statistics Approach

21

7.1 Hotspot Detection The present work addressed the concept of the hotspots of cancer incidences and mortality using scan statistics. We have used the projected population of the year 2020 for hotspot analysis of cancer incidence and mortality. The calculated results are shown in Tables 2, 3, 4, and 5. We calculated the cancer incidence hotspots for the years 2010-2020. The columns of Table 2 show the hotspot region and the value of relative risk, log-likelihood ratio, and p-statistics. The state of Kerala is followed closely by Delhi, Gujarat, Haryana, and Karnataka in a particular order (shown in Table 2). The result shows that the state of Kerala emerged as the primary hotspot with a relative risk value of (158.41), log-likelihood ratio value of 1843307.228884, and p-value of 0.00000000000000001. Figure 4 presents increasing trends of cancer incidence in Kerala. We have computed the coldspot of cancer incidence for the years 2010-2020, as shown by the columns of Table 3, which consists of the coldspot region, relative risk value, log-likelihood ratio value, and p-values, respectively. Goa, with a relative risk value of 0.024, log-likelihood ratio value of 591846.612982, and a p-value of 0.00000000000000001, emerged as the most likely coldspot cluster, followed by Daman and Diu, Jammu and Kashmir, Jharkhand, and Uttar Pradesh, respectively. The computational result of the cancer mortality hotspot for the years 2010-2018 is presented by the column of Table 4. Figure 5 presents increasing trends in cancer mortality in Bihar. It is important to highlight the fact that Bihar (relative risk value of 3.15, log-likelihood ratio value of 219643.412516, and p-value of 0.00000000000000001) has emerged as the cancer mortality hotspot followed by Uttar Pradesh and Telangana. Table 5 presents the results of the cancer mortality coldspot for the years 2010-2018, which shows Kerala, Lakshadweep, Puducherry, Karnataka, Tamil Nadu, and Goa are the primary coldspots. We found it very interesting to present the findings that all these coldspots except Goa belong to the southern states of India, highlighting relatively better health and medical facilities and effective health management in comparison to the country’s other regions and states (India). We must highlight that though Kerala has been the hotspot of cancer incidences, it has also been the coldspot of cancer mortality. The direct implication is that human efforts and a holistic integrated approach could bring credible success. We would like to highlight another interesting finding that though Kerala has been the hotspot of cancer incidence, it is not the hotspot of cancer mortality. On the contrary, Bihar has been the hotspot of cancer mortality but not the hotspot of cancer incidence. This poses a serious threat because Bihar is the third most populated state in India and has very poor records on comparative health infrastructure. For example, Bihar witnessed the lowest per capita health expenditure (Rs.558) which is almost half of the national average (Rs. 1112): The HDI value of Bihar is 0.574 which is below the national average value of 0.646. On the other hand, the HDI value of Kerala is 0.782 which is substantially above the national average value of 0.646. We have shown the hotspots and the coldspot regions for cancer incidence using the generic and colored map of India as shown in Figs. 6, and 7 presents the hotspot and coldspot regions for cancer mortality, respectively. Also, we found that states having poor performances on the

22

S. Kaindal et al.

Table 2 Cancer incidence hotspots during the years 2010–2020 Hotspot region

Relative risk

Log-likelihood ratio

p-value

Kerala

158.41

1,843,307.228884

0.00000000000000001

Delhi

13.16

347,889.302816

0.00000000000000001

Gujarat

2.53

223,988.420055

0.00000000000000001

Haryana

3.95

184,263.382552

0.00000000000000001

Karnataka

2.14

166,624.160198

0.00000000000000001

Table 3 Cancer incidence coldspots during the years 2010–2020 Coldspot region

Relative risk

Log-likelihood ratio

p-value

Goa

0.024

591,846.612982

0.00000000000000001

Daman and Diu

0.015

177,946.643258

0.00000000000000001

Jammu and Kashmir

0.36

107,651.792765

0.00000000000000001

Jharkhand

0.54

88,603.457860

0.00000000000000001

Uttar Pradesh

0.88

16,524.044047

0.00000000000000001

Table 4 Cancer mortality hotspots for the years 2010–2018 Hotspot region

Relative risk

Log-likelihood ratio

p-value

Bihar

3.15

219,643.412516

0.00000000000000001

Uttar Pradesh

1.52

62,501.708992

0.00000000000000001

Telangana

1.78

24,913.272607

0.00000000000000001

Table 5 Cancer mortality coldspot 2010–2018 Coldspot region

Relative risk

Log-likelihood ratio

p-value

Kerala, Lakshadweep, Puducherry, Karnataka, Tamil Nadu, Goa

0.58

106,815.148168

0.00000000000000001

above-mentioned factors may emerge as hotspots of cancer deaths despite being the coldspot of cancer incidences. Table 9 presents the two comparative performances between Kerala and Bihar on various indicators that can be used to validate our findings. For example, Kerala is smaller than Bihar in area as well as population but consists of seventeen hundred fifty-five government schools compared to Bihar, which has one thousand fifty-seven government schools only. Similarly, the literacy rate of Kerala is 93.9 which is far above that of Bihar (63.9). The per capita income in Kerala is Rs.111, 661.333 which is far ahead that of Bihar (Rs. 23314.5). There are 10,404 screened cases in Kerala whereas only 17 screened cases in Bihar in the year 2020.

Cancer Hotspot Identification and Analysis: A Scan Statistics Approach

23

Fig. 4 Trend of cancer incidence in Kerala

Fig. 5 Trend of cancer mortality in Bihar

7.2 Descriptive Approach In the present work, we identified the top five most consistent sites of cancer in both genders for the years 2008, 2009, 2010, 2015, and 2020. Cancer sites, namely eesophagus, myeloid leukemia, rectum, larynx, and brain-NS, were assigned as the top five most consistent cancer sites among males in India. Similarly, the larynx, hypopharynx, mouth, tongue, and lungs are the top five most consistent cancer sites among females. Most consistent cancer sites enable us to develop a strategy that necessitate a comprehensive and integrated approach to eliminating the roots of cancer. This can be considered as a first step toward a cancer eradication programme in our country. Tables 6 and 7 present the above results effectively (in terms of numeric values). The box plot presented in Figs. 2 and 3 represents the most consistent cancer sites among males and females, respectively. The important question that comes to the mind of the researcher is whether there is a significant difference in cancer sites among males and females. It can be claimed that there is no significant difference in cancer sites among the two genders. We attempted to address this issue using a Welch t-test, and the computed result of this test is shown in Table 8. We found sufficient evidence in our sample against the null hypothesis. So, we reject the null hypothesis

24

S. Kaindal et al.

Fig. 6 Hotspot and coldspot regions of cancer incidence

(H 0 ) in favor of the alternative hypothesis (H 1 ). This implies a significant difference in cancer sites between the two genders. We found that there does not exist a strong rank-based correlation among the prevalent cancer sites between the two genders. As such, we may claim man-made reasons responsible for occurrences of different cancer sites among the two genders. As such, we found that cancer incidences and mortality hotspots are not due to chance but due to human adoption of a sedentary lifestyle and deviation from the natural way of life (Table 9).

Cancer Hotspot Identification and Analysis: A Scan Statistics Approach

25

Fig. 7 Hotspot and coldspot regions of cancer mortality

Table 6 Top five cancer sites among males for the years 2008, 2009, 2010, 2015, and 2020 Cancer sites (male)

ICD-10 code

Coefficient of variation

Rank

Esophagus

C15

0.054951

1

Myeloid Leukemia

C92-C94

0.075571

2

Rectum

C19-C20

0.075574

3

Larynx

C32

0.075575

4

Brain NS

C70-C72

0.075575

5

Table 7 Top five cancer sites among females for the years 2008, 2009, 2010, 2015, and 2020

Cancer sites

ICD-10 code

Coefficient of variation

Rank

Larynx

C32

0.08932

1

Hypopharynx

C12-C13

0.089363

2

Mouth

C03-C06

0.089384

3

Tongue

C01-C02

0.089385

4

Lung

C33-C34

0.089392

5

26

S. Kaindal et al.

Table 8 Welch T-test Analysis

Variable1

Variable2

Analysis

Variable1

Mean

23,203.533

14,127.6

T stat

2.5960

Variance

151,192,864.1

32,143,617.97

P(T