Design and Forecasting Models for Disease Management
Description
The book provides an essential overview of AI techniques in disease management and how these computational methods can lead to further innovations in healthcare.
Design and Forecasting Models for Disease Management is a resourceful volume of 13 chapters that elaborates on computational methods and how AI techniques can aid in smart disease management. It contains several statistical and AI techniques that can be used to acquire data on many different diseases. The main objective of this book is to demonstrate how AI techniques work for early disease detection and forecasting useful information for medical experts. As such, this volume intends to serve as a resource to elicit and elaborate on possible intelligent mechanisms for helping detect early signs of diseases. Additionally, the book examines numerous machine learning and data analysis techniques in the biomedical field that are used for detecting and forecasting disease management at the cellular level. It discusses various applications of image segmentation, data analysis techniques, and hybrid machine learning techniques for illnesses, and encompasses modeling, prediction, and diagnosis of disease data.
Audience
Researchers, engineers and graduate students in the fields of computational biology, information technology, bioinformatics, and epidemiology.
More details
Other editions
Additional editions
Persons
Pijush Dutta, PhD, is an assistant professor and head of the Department of Electronics and Communication Engineering at Greater Kolkata College of Engineering and Management, West Bengal, India, with over 11 years of teaching and over seven years of research experience. He has published eight books, as well as 14 patents and over 100 research articles in national and international journals and conferences. His research interests include sensors and transducers, nonlinear process control systems, the Internet of Things (IoT), and machine and deep learning.
Sudip Mandal, PhD, is an assistant professor in the Electronics and Communication Engineering Department at Jalpaiguri Government Engineering College, India. He has over 50 publications in national and international peer-reviewed journals and conferences, as well as two Indian patents and two books. He is a member of the Institute of Electrical and Electronics Engineers' Computational Intelligence Society.
Korhan Cengiz, PhD, is an associate professor in the Department of Computer Engineering at Istinye University, Istanbul, Turkey. He has published over 40 articles in international peer-reviewed journals, five international patents, and edited over ten books. His research interests include wireless sensor networks, wireless communications, and statistical signal processing.
Arindam Sadhu, PhD, is an assistant professor in the Electronics and Communication Engineering Department at Swami Vivekananda University, West Bengal, India, with over five years of teaching and over three years of research experience. He has published two international patents and over ten articles in national and international journals and conferences. His research interests include post-complementary metal-oxide-semiconductor transistors, quantum computing, and quantum dot cellular automata.
Gour Gopal Jana is an assistant professor in the Electronics and Communication Engineering Department at Greater Kolkata College of Engineering and Management, West Bengal, India, with over 13 years of teaching and over three years of research experience. He has published two international patents and over ten research articles in national and international journals and conference proceedings. His research interests include metal thin film sensors, biosensors, nanobiosensors, and nanocomposites.
Content
Preface xvii
Part 1: Safety and Regulatory Aspects for Disease Pre-Screening 1
1 A Study of Possible AI Aversion in Healthcare Consumers 3
Tanupriya Mukherjee and Anusriya Mukherjee
1.1 Introduction to AI in Healthcare 4
1.1.1 The Role of AI in Transforming Healthcare 5
1.1.2 The Unfolding Paradigm: Potential Benefits and Challenges of AI Implementation in Healthcare 6
1.1.3 Overview of Consumer Receptivity Towards AI in Medicine: A Comparative Analysis 7
1.2 Consumer Reluctance to Utilize AI in Healthcare: Present Scenario 8
1.2.1 Top Factors Influencing Consumer Resistance to Medical AI 10
1.2.2 Uncovering the Psychological Barriers and Concerns Associated with AI Adoption in Healthcare 11
1.2.3 Case Studies and Research Findings on Consumer Aversion to AI-Based Healthcare Services 13
1.2.4 Impact on Consumer Decision-Making 14
1.2.5 Effects of AI Aversion on Consumer Decision-Making Processes: An Analysis 15
1.2.6 Understanding How Consumer Perceptions Influence Their Choice Between Human and AI Healthcare Providers 15
1.2.7 Exploring Role of Trust, Perceived Competence and Empathy in Consumer Preferences 16
1.3 Economic Implications of AI Aversion 17
1.3.1 Investigating Influence of AI Aversion on Consumer Willingness to Pay for Healthcare Services 19
1.3.2 Influence of Patient Education on AI Aversion in Healthcare 19
1.3.3 Influence of Patient Awareness on AI Aversion in Healthcare 21
1.3.4 Influence of Age of Patient on AI Aversion in Healthcare 21
1.4 Overcoming Resistance to Medical AI 22
1.4.1 Strategies for Enhancing Consumer Trust and Acceptance of AI in Healthcare 23
1.4.2 Approaches to Alleviate Consumer Concerns and Misconceptions: Communication and Education 24
1.4.3 Cases of Successful Implementation of AI Technologies in Healthcare and Lessons Learned 25
1.5 Ethical Considerations and Governance 26
1.5.1 Regulatory Frameworks for Ethical AI Operations to Fight Aversion in Healthcare Consumers 27
1.5.2 Addressing the Potential Cost-Effectiveness and Affordability Concerns Associated with AI-Based Healthcare Solutions 28
1.5.3 Balancing Privacy, Data Protection and Need for Transparency in AI Healthcare Applications 29
1.6 Future Outlook and Opportunities 31
1.6.1 The Future of AI in Healthcare and Its Impact on Consumer Aversion 32
1.6.2 Exploring Emerging Technologies and Trends That May Alleviate Consumer Concerns 33
1.6.3 Opportunities for Collaboration Between AI Developers, Healthcare Providers, and Consumers 34
1.6.4 Summary of Key Findings on Consumer Aversion to AI in Healthcare 35
1.6.5 Implications for Healthcare Practitioners, Policymakers and Researchers 36
1.7 Conclusion 37
References 38
2 A Study of AI Application Through Integrated and Systematic Moral Cognitive Therapy in the Healthcare Sector 47
Anusriya Mukherjee, Tanupriya Mukherjee and Mili Mitra Roy
2.1 Introduction 48
2.1.1 Understanding the Role of AI in Healthcare 49
2.1.2 Advantages of AI in Healthcare 50
2.1.3 Moral Dilemmas and AI-Based Healthcare 52
2.2 What is Integrated and Systematic Moral Cognitive Therapy (ISMCT)? 54
2.2.1 Integrating Moral Cognitive Therapy with AI 55
2.2.2 Alignment of Moral Cognitive Therapy Principles with AI Applications 56
2.2.3 Benefits of Integrated and Systematic Moral Cognitive Therapy 57
2.2.4 Applications of AI-Integrated Moral Cognitive Therapy in Healthcare 58
2.3 The Role of AI in Healthcare: A Fine Balance Between Ethics and Innovation 61
2.3.1 Humanizing Healthcare: Towards an AI-ISMCT 62
2.3.2 Synergized AI and ISMCT 63
2.3.3 Case Study and Success Stories 64
2.4 Advancing Research in AI-Integrated Moral Cognitive Therapy 67
2.4.1 Collaborative Efforts Between Healthcare Professionals and AI Developers 68
2.4.2 Implications for Policy and Regulatory Frameworks 69
2.5 Conclusion 70
References 70
3 A Strategic Model to Control Non-Communicable Diseases 77
Soumik Gangopadhyay, Amitava Ukil, Soma Sur and Saugat Ghosh
3.1 Introduction 78
3.1.1 India and NCDs 78
3.2 Survey of Literature 84
3.2.1 Factors Contributing to the Growth of NCDs 84
3.2.2 Lifestyle Modification - A Strategic Role in Mitigation of NCD 85
3.2.3 Policy to Control NCDs 86
3.3 Proposed Model 87
3.3.1 Registration and Information Centre (RIC) 88
3.3.2 Integration Centre (IIC) 88
3.3.3 Strategic Review Centre (SRC) 89
3.3.4 Expected Outcome of the Proposed Model 90
3.4 Conclusion 91
References 92
4 Image Compression Technique Using Color Filter Array (CFA) for Disease Diagnosis and Treatment 99
Indrani Dalui, Avisek Chatterjee, Surajit Goon and Pubali Das Sarkar
4.1 Introduction 100
4.1.1 Color Filter Array 100
4.1.2 Electronic Health Record (EHR) 101
4.2 Related Works 102
4.3 Proposed Model 108
4.4 Implementation 110
4.5 Results 111
4.6 Conclusion 112
References 113
5 Research in Image Processing for Medical Applications Using the Secure Smart Healthcare Technique 115
Debraj Modak and Chowdhury Jaminur Rahaman
5.1 Introduction 116
5.1.1 Imaging Systems 118
5.1.2 The Digital Image Processing System 119
5.1.3 Image Enhancement 120
5.2 Classification of Digital Images 121
5.2.1 Utilizations of Digital Image Processing (DIP) 121
5.2.1.1 Medicine 121
5.2.1.2 Forensics 122
5.2.2 Medical Image Analysis 122
5.2.3 Max-Variance Automatic Cut-Off Method 122
5.2.4 Medical Imaging Segmentation 124
5.2.5 Image-Based on Edge Detection 124
5.2.5.1 Robert's Kernel Method 125
5.2.5.2 Prewitt Kernel 125
5.2.5.3 Sobel Kernel 125
5.2.5.4 k-Means Segmentation 126
5.2.6 Images from ¿-Rays 126
5.2.6.1 Non-Ionizing Radiation 127
5.2.6.2 Magnetic Resonance Imaging 128
5.2.6.3 Segmentation Using Multiple Images Acquired by Different Imaging Techniques 129
5.3 Methods 130
5.3.1 k-Means Approach 130
5.3.2 Bayesian Objective Function 132
5.4 Segmentation and Database Extraction with Neural Networks 133
5.4.1 Artificial Neural Network 133
5.4.2 Bayesian Belief Networks 134
5.5 Applications in Medical Image Analysis 135
5.5.1 Using Artificial Neural Network for Better Optimization and Detection in Medical Imaging 136
5.5.1.1 Opportunities 136
5.6 Standardize Analytics Pipeline for the Health Sector 136
5.7 Feature Extraction/Selection 138
5.7.1 Significance of Machine Learning for Medical Image Processing 138
5.7.2 Significance of Deep Learning for Medical Image Processing 139
5.8 Image-Based Forecasting Using Internet of Things (IoT) in Smart Healthcare System 141
5.9 IoT Monitoring Applications Based on Image Processing 143
5.10 Significance of Computer-aided Big Healthcare Data (BHD) for Medical Image Processing 145
5.11 Applications of Big Data 147
5.11.1 Big Data Analytics in Health Sector 147
5.11.2 Computer-Aided Diagnosis in Mammography 149
5.11.3 Tumor Imaging and Treatment 149
5.11.4 Molecular Imaging 149
5.11.5 Surgical Interventions 150
5.12 Conclusion 150
References 151
6 Comparative Study on Image Enhancement Techniques for Biomedical Images 155
Sudip Mandal, Uma Biswas, Aparna Mahato and Aurgha Karmakar
6.1 Introduction 156
6.2 Literature Review 157
6.3 Theoretical Concepts 158
6.3.1 Logarithmic Transformation 159
6.3.1.1 Advantages of Log Transformation 160
6.3.1.2 Limitations of Log Transformation 160
6.3.2 Power Law Transformation or Gamma Correction 160
6.3.2.1 Advantages of Gamma Correction 161
6.3.2.2 Limitations of Gamma Correction 161
6.3.3 Piecewise Linear Transformation or Contrast Stretching 162
6.3.3.1 Advantages of Contrast Stretching 162
6.3.3.2 Limitations of Contrast Stretching 163
6.3.4 Histogram Equalization 163
6.3.4.1 Advantages of Histogram Equalization 164
6.3.4.2 Limitations of Histogram Equalization 164
6.3.5 Contrast-Limited Adaptive Histogram Equalization (clahe) 164
6.3.5.1 Advantages of CLAHE 165
6.3.5.2 Limitation of CLAHE 165
6.3.6 Adjustment Function 166
6.4 Results and Discussion 166
6.4.1 Images and Histograms for Different Images Using Different Enhancement Methods 167
6.4.2 Comparison for Different Image Enhancement Techniques 175
6.5 Conclusion 178
References 179
7 Exploring Parkinson's Disease Progression and Patient Variability: Insights from Clinical and Molecular Data Analysis 181
Amit Kumar, Neha Sharma and Korhan Cengiz
7.1 Introduction 182
7.2 Literature Review 183
7.3 Data Review 184
7.3.1 Clinical Data 185
7.3.2 Peptides Data 192
7.3.3 Protein Data 194
7.4 Parkinson's Dynamic for Patients in Train 196
7.5 Conclusion 197
References 198
8 A Survey-Based Comparative Study on Machine Learning Techniques for Early Detection of Mental Illness 201
Prachi Majumder, Sompadma Mukherjee, Shreyashi Saha, Tamasree Biswas, Mousumi Saha, Deepanwita Das and Suchismita Maiti
8.1 Introduction 201
8.2 Background 202
8.3 Review of Previous Works 203
8.3.1 Standard Questionnaire 203
8.3.2 Social Media Content 206
8.4 Comparative Result 208
8.5 Discussion 212
8.6 Conclusion 213
References 213
Part 2: Clinical Decision Support System for Early Disease Detection and Management 215
9 Diagnostics and Classification of Alzheimer's Diseases Using Improved Deep Learning Architectures 217
Mainak Dey, Pijush Dutta and Gour Gopal Jana
9.1 Introduction 218
9.2 Related Works 219
9.3 Method 222
9.3.1 Data Description 224
9.4 Result Analysis 225
9.4.1 Performance Metrics 227
9.4.2 Experimental Setup 230
9.5 Conclusion 232
Data Availability 233
References 233
10 Perform a Comparative Study Based on Conventional Machine Learning Approaches for Human Stress Level Detection 237
Pratham Sharma, Prerana Singh, Mahe Parah, Shyamapriya Chatterjee, Anirban Bhar, Soumya Bhattacharyya and Pijush Dutta
10.1 Introduction 238
10.2 Related Work 239
10.3 Architecture Design 242
10.3.1 Body Temperature 243
10.3.2 Humidity Analysis 243
10.3.3 Step Count Analysis 243
10.3.4 Dataset 243
10.4 Experiment 244
10.4.1 Performance Matrices 245
10.5 Result Analysis 246
10.6 Conclusion 248
References 249
11 Diabetes Prediction Using a Hybrid PCA-Based Feature Selection and Computational Machine Learning Algorithm 253
Sumanta Dey, Pijush Dutta, Gour Gopal Jana and Arindam Sadhu
11.1 Introduction 254
11.2 Related Work 254
11.3 Proposed Workflow 256
11.3.1 Data Pre-Processing 256
11.3.2 Feature Selection 257
11.3.3 Dimensionality Reduction 258
11.3.4 Classification 259
11.4 Result Analysis 261
11.4.1 Evaluation Criteria 261
11.5 Conclusion and Future Work 265
References 266
12 A Robust IoT-Based Approach to Enhance Cybersecurity and Patient Trust in the Smart Health Care System: Zero-Trust Model 269
Raghunath Maji, Biswajit Gayen and Sandeepan Saha
12.1 Introduction 270
12.2 Security Threats on Smart Healthcare 271
12.2.1 Medical Data Monitoring and Patient Privacy Information 271
12.2.2 Network Attacks on Critical Infrastructures 272
12.2.3 Malicious Data Tampering 272
12.3 Smart Healthcare Security and Four-Dimension Model 273
12.3.1 Subject 273
12.3.2 Object 274
12.3.3 Environment 275
12.3.4 Behavior 275
12.3.5 Risk Assessment and Security Checking 275
12.4 Conclusion and Future Prospects 279
Acknowledgment 280
References 280
13 Safeguarding Digital Health: A Novel Approach to Malicious Device Detection in Smart Healthcare 283
Raghunath Maji and Biswajit Gayen
13.1 Introduction 284
13.2 Related Work 286
13.3 Our Proposed Framework 289
13.4 Overview of Our Proposed Framework 289
13.5 Evaluation Procedure 291
13.6 Performance Evaluation 292
13.7 Conclusion 293
References 294
Index 297
Preface
Overview
The novel application of artificial intelligence (AI) techniques, including machine learning, and deep learning analytics for design and forecasting models for disease management, is an emerging area in computer science, computation biology, information technology, bioinformatics, bioinformatics, and epidemiology. During the last 20 years, various AI approaches have been successfully applied in diverse fields. Medicine is one of the most active domains for AI techniques, since they facilitate model diagnostics information based on statistical data and reveal hidden dependencies between symptoms of major diseases.
The present capacity to develop, evaluate, manufacture, distribute, and administer effective medical countermeasures is inadequate to meet the burden of both recurrent and emerging outbreaks of infectious diseases. When such interventions are unavailable, public health measures and supportive clinical care remain the only feasible tools to slow an emerging outbreak. Decision-making under such circumstances can be greatly improved by the use of appropriate data and advanced analytics, such as infectious disease modeling. Furthermore, these analyses can guide decision-making when medical countermeasures become available, allowing them to be used more effectively.
Forecasting is an emerging analytical capability that has demonstrated value in recent outbreaks by informing policy and enabling real-time epidemic management decisions. Since healthcare is one of the key parameters in assessing the gross domestic product (GDP) of any country, it has become crucial to transition from traditional healthcare practices to a smart healthcare system. New healthcare technologies provide numerous opportunities to maximize disease recognition, analysis, and other natural variables that may affect it. Therefore, it is necessary to understand how computer-assisted technologies can best be utilized and adopted in the conversion to smart healthcare.
This book is an essential publication that widens the spectrum of computational methods that can aid in smart disease management. It contains several statistical and AI techniques that can be used to acquire data of many different diseases.
Objective
AI techniques for early disease detection and forecasting can reveal useful information to medical experts. As such, this volume intends to serve as a resource to elicit and elaborate on possible intelligent mechanisms for helping in the modeling, prediction, and diagnosis of the infected individuals, as well as providing means for detecting early signs of the disease. As such, this book examines numerous ML and DL techniques in the biomedical field that are used for detecting and forecasting disease management at the cellular level. It discusses applications for image segmentation, classification, neural networks, and a variety of computational intelligence approaches. It also discusses DL techniques for noise elimination and filtering of brain disease and hybrid ML techniques for diabetes prediction, mental illness, and stress level monitoring approaches.
We hope this book encourages an even wider adoption of these concepts and methods to assist physicians in problem solving and stimulates research that will lead to additional innovations in this area.
Organization
The volume comprises thirteen comprehensive chapters that encompass the modelling, prediction, and diagnosis of disease data. The arrangement of the chapters guides the reader from effective modelling of different disease data to detection and diagnosis. Chapter 1 presents an overview of how AI is transforming healthcare, as well as examples of consumer resistance to it, the economic implications of AI aversion, and ethical consideration and governance. Hand in hand with the emerging technology, healthcare systems are becoming progressively complex while constantly adapting to socioeconomic, epidemiological, and demographic changes, contributing to intricate global healthcare objectives.
Chapter 2 explores the use of integrated, and systematic moral cognitive therapy as a means of administering AI in the healthcare sector. The integration of AI in healthcare has the potential to improve prognosis with the help of an enormous database, but concerns have arisen regarding the vulnerability of a large section of end-users and the ethical implications of its use.
Social inequities, poverty lack of education, and risk of lifestyle diseases are mutually complementary to each other. An unhealthy lifestyle claims maximum responsibility among the factors of becoming a victim of lifestyle diseases. Zero preventive options other than lifestyle modification have enhanced the complication of management of such mass health issues.
Chapter 3 investigates a conceptual model that can bridge the gaps between healthcare, policy, and disease prevention. Such a model can assist with the recording and reporting of data to analyze the victim's pattern of lifestyle adoption in place of the current healthcare alternative. Color Filter Array (CFA) data can be efficiently compressed using existing image compression algorithms such as JPEG. In addition, it tested the overall performance of CFA compression methods and demosaicing algorithm selection. These CFA compression methods exploit how CFA images do not contain repetitive data introduced by demosaicing.
Chapter 4 establishes a relationship between the discovery of different CFA compression techniques and describes how the choice of demosaicing algorithm affects visual quality. Using a simple bilinear demosaicing method, CFA image compression produces better images compared to the traditional compression scheme.
Through image processing techniques, digital images can be compressed, monitored, and transmitted effectively in remote settings. Digital images have a significant effect on modern society, science, technology, and art. Some techniques utilized in image enhancement include filtering, segmentation, object recognition, and image fusion. Image processing techniques have played a crucial role in advancing healthcare, particularly in the field of smart healthcare. Chapter 5 reviews image processing for medical applications using secure smart healthcare techniques. Furthermore, the various challenges that scientists must resolve because of disease are also discussed. Image enhancement techniques engage an essential and fundamental task and interpretability of biomedical images, enabling accurate diagnosis and analysis. This chapter provides a comprehensive overview of image enhancement techniques that are specifically tailored for biomedical images. Chapter 6 presents various image enhancement methods designed specifically for biomedical imaging. The chapter evaluates and contrasts the usefulness of various approaches and the extent to which they are beneficial to the biomedical field.
Disease progression prediction is essential for patient care and therapeutic development. Clinical, genetic, and molecular data related to Parkinson's disease (PD) are collectively gathered by the Accelerating Medicines Partnership (AMP). Chapter 7 presents parameters linked to Parkinson's disease progression and evaluates the appropriateness of the AMP PD dataset for predictive modeling through an Exploratory Data Analysis (EDA). The onset of the Covid-19 pandemic and the practice of social isolation have led to a rise in instances of mental health issues. Advances in machine learning in the present decade have created additional opportunities for detection and the identification of mental health issues. Chapter 8 investigates the automatic prediction and diagnosis of mental illness using machine learning algorithms.
One of the most common neurodegenerative illnesses that affect older adults and mostly impair memory in the brain is Alzheimer's disease. Chapter 9 presents three models, efficient net B2, efficient net B3, and efficient net B4 architecture-based deep learning solutions for detecting and categorizing Alzheimer's disease.
One of the biggest issues in today's world is mental stress. Now that age is not taken into account when calculating stress levels, it makes no difference. Depression, heart attacks, and even suicide can result from extreme stress. Stress may have an impact on many facets of our existence, such as our behavior, emotions, and capacity for thought. Chapter 10 presents a machine-learning approach for measuring a person's stress level using three crucial parameters: body temperature, step count, and humidity.
Diabetes Mellitus is commonly known as a metabolic issue in which the body cannot utilize insulin, store glucose for energy, or produce insulin. Diabetes patients can suffer various sicknesses that incorporate kidney disappointment, stroke, visual impairment, cardiovascular failures, and lower appendage removal. Chapter 11 presents a hybrid-based machine learning approach for identifying the potential chances of contracting diabetes diseases.
The increasing adoption of smart healthcare care and Internet of Things (IoT) devices has revolutionized the healthcare industries, offering enhanced patient monitoring, diagnosis, and treatment. However, the integration of these devices into the healthcare ecosystem raises serious concerns regarding the security and privacy of sensitive medical records. Chapter 12 contributes to strengthening the overall cyber security framework within the healthcare industry, ensuring patients' trust in smart healthcare IoT devices and fostering the continued growth of remote patient monitoring.
The proliferation...
