Computational Intelligence in Oncology: Applications in Diagnosis, Prognosis and Therapeutics of Cancers (Studies in Computational Intelligence Book 1016)

This book encapsulates recent applications of CI methods in the field of computational oncology, especially cancer diagnosis, prognosis, and its optimized therapeutics. The cancer has been known as a heterogeneous disease categorized in several different subtypes. According to WHO’s recent report, cancer is a leading cause of death worldwide, accounting for over 10 million deaths in the year 2020. Therefore, its early diagnosis, prognosis, and classification to a subtype have become necessary as it facilitates the subsequent clinical management and therapeutics plan. Computational intelligence (CI) methods, including artificial neural networks (ANNs), fuzzy logic, evolutionary computations, various machine learning and deep learning, and nature-inspired algorithms, have been widely utilized in various aspects of oncology research, viz. diagnosis, prognosis, therapeutics, and optimized clinical management. Appreciable progress has been made toward the understanding the hallmarks of cancer development, progression, and its effective therapeutics. However, notwithstanding the extrinsic and intrinsic factors which lead to drastic increment in incidence cases, the detection, diagnosis, prognosis, and therapeutics remain an apex challenge for the medical fraternity. With the advent in CI-based approaches, including nature-inspired techniques, and availability of clinical data from various high-throughput experiments, medical consultants, researchers, and oncologists have seen a hope to devise and employ CI in various aspects of oncology. The main aim of the book is to occupy state-of-the-art applications of CI methods which have been derived from core computer sciences to back medical oncology. This edited book covers artificial neural networks, fuzzy logic and fuzzy inference systems, evolutionary algorithms, various nature-inspired algorithms, and hybrid intelligent systems which are widely appreciated for the diagnosis, prognosis, and optimization of therapeutics of various cancers. Besides, this book also covers multi-omics exploration, gene expression analysis, gene signature identification of cancers, genomic characterization of tumors, anti-cancer drug design and discovery, drug response prediction by means of CI, and applications of IoT, IoMT, and blockchain technology in cancer research. Preface About This Book Contents Editor and Contributors Preliminaries Computational Intelligence in Oncology: Past, Present, and Future 1 Introduction 1.1 From AI to Computational Intelligence: The Transition 1.2 Motivation: CI in Oncology 2 Oncology: Origin, Evolution, Current Status, and Future 2.1 Origin 2.2 Evolution 2.3 Current Status and Future 3 Computational Oncology 3.1 Background and History 3.2 Data Analysis and Applications 4 How CI is Strengthening Oncology Research? 4.1 Classification 4.2 Imaging Identification and Analysis 4.3 Risk Assessment 4.4 Cancer Therapy 4.5 Cancer Treatment Optimization 4.6 Diagnosis and Screening in Radiology 5 Promises and Challenges of CI in Oncology 6 Discussion and Conclusion References Machine Learning-Based Models in the Diagnosis, Prognosis and Effective Cancer Therapeutics: Current State-of-the-Art 1 Introduction 2 Machine Learning in Prediction and Prognosis of Cancer 2.1 Cancer Susceptibility Prediction 2.2 Cancer Recurrence Prediction 2.3 Cancer Survival Prediction 3 Chronological Review 4 Machine Learning-Based Cancer Diagnosis System: A General Pipeline 5 Models for Detection of Various Cancers Types 5.1 Breast Cancer 5.2 Lung Cancer 5.3 Brain Cancer 5.4 Skin Cancer 5.5 Prostate Cancer 6 Models for Predicting Toxicity in Various Cancerous Types 6.1 Brain 6.2 Breast 6.3 Esophagus 6.4 Gynecological Cancers 6.5 Head and Neck 6.6 Liver 6.7 Lung 6.8 Prostate 7 Challenges and Future of Machine Learning in Cancer Research 8 Conclusion References Computational Intelligent Systems in Oncology: A Way Toward Translational Healthcare 1 Introduction 2 Rise of Computational Intelligent Systems in Oncology 3 Transition from Traditional Approaches to CIS 4 CIS in Translational Healthcare 5 Challenges and Future Prospects of CIS in Oncology 6 Conclusions References Computational Resources for Oncology Research: A Comprehensive Analysis 1 Introduction 2 Computational Resources in Oncological Research 2.1 Resources for Cancer Genomics Study 2.2 Bioinformatics Tools for Cancer Metabolomics 2.3 Tools and Web Servers for Cancer Prognosis Analysis 2.4 Servers and Tools for Cancer Biomarker Development 2.5 Software Tools for Cancer Hazard Identification 2.6 Web-Based Tools/Databases for Analyzing TCGA Data 2.7 Other’s Databases/Tools/Webservers in Oncological Research to Identify Gene Function and Structure Prediction 3 Cancer Precision Drug Discovery: Computational Biology and AI 4 Limitations and Prospects 4.1 Terminology Inconsistency 4.2 Lack of Interoperability 4.3 Database Proprietary Issues 4.4 Payment Issues 4.5 Inadequate Support 4.6 Misleading Results 4.7 Algorithms 5 Conclusions References Cancer Detection, Diagnosis, Survival, and Recurrence Prediction Application of Convolutional Neural Networks in Cancer Diagnosis 1 Introduction 1.1 Artificial Neural Network 1.2 Convolutional Neural Network 2 Architecture of a CNN 2.1 Convolutional Layer 2.2 Pooling Layer 2.3 Activation Function 2.4 Fully Connected Layers 2.5 Weights and Biases 2.6 Dropouts 2.7 Batch Normalization 3 Data Source, Preprocessing, and Model Performance 3.1 Types of Medical Imaging for Cancer Diagnosis and Treatment 3.2 Open Cancer Datasets 3.3 Pre-processing of Images 3.4 Evaluating the Performance of a Model 4 Recent Advances in CNN Models for Cancer Diagnosis 4.1 Breast Cancer Diagnosis 4.2 Lung Cancer 4.3 Skin Cancer 4.4 Multiple Myeloma 5 Limitations and Challenges 6 Conclusion References Automatic Cancer Detection Using Probabilistic Convergence Theory 1 Introduction 2 Literature Survey 3 Proposed Work 4 Edge Detection Method—“Sobel Operator” 4.1 Edge Detection Method—“Laplacian” 4.2 Intermediate Laplacian Matrix 5 Conclusion 5.1 Future Prospects References Computational Intelligence Methods for Cancer Survival Prediction 1 Introduction 2 Survival Models in Predicting Outcomes 3 Survival Model Analysis: Basic Concepts 4 CI Methods Utilized in Survival Prediction: Concept Examples 5 Classification of CI Methods and Application Challenges Based on Survival Dataset 6 CI Methods for Survival Prediction Analysis 6.1 Survival Trees Method 6.2 Bayesian Methods 6.3 Artificial Neural Networks Method 6.4 Support Vector Machines Method 7 CI Techniques Opportunities and Future Direction Toward Survival Prediction 8 Conclusions References Breast Cancer Survival Prediction Using Machine Learning 1 Introduction 2 Methodology and Data Acquisition 2.1 Data Comprehension 2.2 Preparation of Data 2.3 Machine Learning Analysis 2.4 Evaluation of the Model 3 Results 4 Discussion 5 Conclusion References Deep Learning Models for Classification of Brain Tumor with Magnetic Resonance Imaging Images Dataset 1 Introduction 2 Deep Learning 2.1 ResNet50 2.2 VGG16 3 Related Work 4 Methods and Material 4.1 Dataset 4.2 Dataset Argumentation 4.3 Environmental Setup 5 Deep Learning Models for Classification of Brain Tumor 6 Performance Evaluation of Deep Learning Models for Classification of Brain Tumor 7 Future Prospects and Current Limitations 8 Conclusion References Predicting the Cancer Recurrence Using Artificial Neural Networks 1 Introduction 2 Artificial Neural Networks 2.1 Convolutional Neural Networks 2.2 Periodic Neural Networks 2.3 Modular Interconnected Network 3 ANN in Breast, Ovarian, and Lung Cancer 3.1 Breast Cancer 3.2 Ovarian Cancer 3.3 Lung Cancer 4 ANNs in Cancer Recurrence Prediction 5 Discussion 6 Conclusion References Computer Intelligence in Detection of Malignant or Premalignant Oral Lesions: The Story So Far 1 Introduction 1.1 Computer Tomography (CT) Scan 1.2 Positron Emission Tomography (PET) Scan 1.3 Magnetic Resonance Imaging (MRI) 1.4 Endoscopy 2 Computational Intelligence and Oral Cancer 3 Computational Intelligence and the Patient Perspective 4 Computational Intelligence and the Patient Workup Plan 5 Computational Intelligence and the Future 6 Conclusion References Fuzzy Logic-Based Hybrid Models for Clinical Decision Support Systems in Cancer 1 Introduction 2 Background of Fuzzy Logic 3 Related Work 4 Limitations of FL-Based CDSS 5 Future Prospects of Fuzzy Logic Systems in CDSS 6 Conclusion References Predicting Cancer Biomarkers, Therapeutic Targets, Drug Response, and Drug Design, Discovery, and Development Predicting Biomarkers and Therapeutic Targets in Cancer 1 Introduction 2 MiRNA Databases 2.1 MicroRNAs as Predictive Biomarkers 2.2 Anomalous Expression of MiRNA in Cancer 3 Generic Pipeline for Biomarkers Detection and Various State-of-the-Art Methods 4 Future Markers 5 Conclusions References Computational Intelligence: A Step Forward in Cancer Biomarker Discovery and Therapeutic Target Prediction 1 Introduction 2 CI Impact of Precision Oncology in the Healthcare Industry 3 NGS-Aiding Molecular Profiling 4 Biomarkers for Detecting the Onset of Disease, Its Diagnosis, and Forecasting the Prognosis 5 Breakthrough of Deep Learning 6 CI in Cancer Imaging and Diagnosis 6.1 Radiographic Imaging 6.2 Digital Pathology 7 Computational Intelligence and Translational Oncology 8 Cancer Therapy: The Quantum Leap 9 Limitations in Medical Image Analysis and Their Proposed Solution 10 Challenges and Future Aspects 10.1 A Unique Data Challenges 10.2 Data Management, Annotation, and Storage 11 Conclusion References Computational Intelligence-Based Cheminformatics Model as Cancer Therapeutics 1 Introduction 2 Structure-Based Cheminformatics Model for Cancer 2.1 Molecular Docking and Molecular Dynamics Simulation 2.2 Cheminformatics 2.3 Proteo-chemometrics 2.4 Molecular Dynamics Simulation 3 Ligand-Based Cheminformatics Cancer Models 4 Random Forest-Based Cancer Model 5 Deep Learning Cancer Models 6 Cancer-Based Phenotypic Models 7 Electronic Health Records-Based ML Methods 8 Conclusion References Gene Expression Signature: An Influential Access to Drug Discovery in Ovarian Cancer 1 Introduction 2 Progression and Development in Ovarian Cancer 3 Cancer Biomarkers 4 Gene Expression Signature and Its Role in Complex Diseases 5 Prognostic Relevance of GES in Ovarian Cancer 6 Gene Expression as a Drug Discovery Tool 7 Conclusion References Machine Learning-Based Approach for Early Diagnosis of Breast Cancer Using Biomarkers and Gene Expression Profiles 1 Introduction 1.1 Biomarkers and Their Role in Breast Cancer Diagnosis 1.2 Gene Expression Profiles 1.3 Why Early Diagnosis of Breast Cancer? 2 Biomarkers and Gene Expression Databases for Breast Cancer 2.1 Subtypes of Biomarkers 2.2 Biomarkers and Gene Expression Databases 2.3 Digital Biomarkers: The Future 3 Computational Intelligence System for Early Diagnosis of Cancer 4 Emerging Machine Learning Methods for Breast Cancer Diagnosis 5 Discussion and Future Perspectives 6 Conclusion References Exosomes: Supramolecular Biomarker Conduit in Cancer 1 Introduction 2 Exosomes 3 Formation and Composition of Exosomes 4 Signalings in the Regulation of Early Endosome 5 Signaling in the Regulation of Exosome Cargo Loading to Late Exosomes 6 Signaling in the Regulation of Exosome Release 7 Exosomes: Specific Purpose 8 Biological Functions of Exosomes in Cancer 9 Isolation of Exosomes 10 Physical Characterization and Molecular Analysis Techniques for Exosomes 11 Exosome-Derived MiRNAs: Cancer Biomarkers 12 Relationship Between Exosome and Tumor Microenvironment 12.1 Exosome and Cancer-Associated Fibroblasts 12.2 Exosomes and Cancer Stem Cells (CSCs) 12.3 Exosome and Mesenchymal Stem Cells (MSCs) 12.4 Exosome and Tumor Microenvironmental Immune Cells 13 Exosomal Diagnostics 13.1 Exosomal Proteins 13.2 Exosomal Nucleic Acids 13.3 Exosomes from Other Biofluids 14 Role of Computation in the Age of Cancer Multi-omics 14.1 Data Acquisition and Processing 14.2 Data Management 15 AI in Cancer Medical Imaging 15.1 Radiographic Imaging 15.2 Digital Pathology 16 AI and Translational Oncology 16.1 Cancer Therapy 16.2 Drug Discovery 17 Machine Learning: Biomarker Identification in Cancer Research 18 Exosome Simplification in Metastatic Cancer Therapy 19 Conclusion References The Revelation and Therapeutic Role of Medicinal Phytochemicals in the Treatment of Cancer: A Brief Review 1 Introduction 2 Phytochemical-Based Treatment of Tumor and Cancer 2.1 Cancer, Tumor Cell Processing 2.2 Looking Back of Phytochemical 2.3 Phytochemical-Based Anticancer Drugs 2.4 Leading Role of the Phytochemicals 2.5 Phytochemicals as a Source of Anticancer Agents 3 Conclusion References Drug Response Prediction Using Machine Learning 1 Introduction 2 ML and AI Tools for Drug Response Prediction 2.1 CancerDP 2.2 Cancer Drug Response Profile Scan (CDRscan) 2.3 DIGREM 2.4 SynergyFinder 2.5 Kernelized Rank Learning 3 Case Study: Development of Genotoxicity Drug Response Prediction Using Machine Learning 3.1 Introduction 3.2 Motivation 3.3 Methodology 3.4 Model Selection, Training, Testing, and Prediction 4 Receiver Operator Characteristic Curve for Model Evaluation 5 Conclusion References The Incipient Role of Computational Intelligence in Oncology: Drug Designing, Discovery, and Development 1 Introduction 2 Conventional Oncology Drug Design, Discovery and Development 3 Challenges in the Drug Development Pipeline 4 Historical Progression of CI 5 Applications of CI in Pharmaceutical Drug Discovery and Development Process 6 Incipient Role of CI in Oncology Drug Discovery 7 CI Platforms Implementation in Drug Designing Pipeline 8 Challenges Faced by CI 9 Concluding Remarks References Gene Expression, Multi-omics Analysis, and Blockchain Computational Intelligence-Based Gene Expression Analysis in Colorectal Cancer: A Review 1 Introduction 2 Computational Intelligence-Based Models: Application in CRC 2.1 Artificial Intelligence 2.2 Machine Learning 2.3 Artificial Neural Networks 3 Significance of Gene Expression Profiling in Cancer Research 4 Gene Expression Classification: A Framework 5 AI and Machine Learning as a Game-Changer in Identifying the Gene Signatures in CRC 5.1 Applications of AI: Cancer Screening 5.2 Applications of Artificial Intelligence: Biomarker Discovery 6 Case Studies 7 Important Advantages of Computational Intelligence 8 Conclusion and Future Scope References Multi-omic Approaches to Improve Cancer Diagnosis, Prognosis, and Therapeutics 1 Introduction 2 Multi-omics Paradigm 3 Methods for Data Integration to Improve Cancer Diagnosis, Prognosis, and Therapeutics 3.1 Multivariate Methods 3.2 Statistically-Based Methods 3.3 Network-Based Methods 3.4 Fusion-Based Methods 3.5 Similarity-Based Methods 3.6 Correlation-Based Methods 3.7 Machine Learning Model 4 Methods for the Detection of Driver Genomic Mutations and Cancer Biomarkers 5 Clinical Translation of “Single Omics” Techniques in Cancer 5.1 Genomics 5.2 Transcriptomics 5.3 Proteomics 5.4 Metabolomics 5.5 Microbiomics 5.6 Multi-omics Data Integration in Cancer Research 6 Clinical Applications of Multi-omics 6.1 Multi-omics Can Diagnose the Earlier Undiagnosed Patients with Unusual Phenotypes 6.2 Pathogenic and Prognostic Biomarkers Can Be Detected by Integrated Multiple Omic Data Sets 6.3 Assisting Early Diagnosis of Cancer 6.4 Both Novel Treatment and Drug Re-Purposing Opportunities Can Be Identified Through Multi-omic Analysis 7 Challenges and Future Perspectives in Multi-omics Data Integration 8 Conclusion References Computational Intelligence Methods for Predicting Cancer Susceptibility from SNP Data 1 Introduction 2 SNPs and Cancer Susceptibility 2.1 Gene Promoter Region 2.2 Exonal SNPs 2.3 Intronal SNPs 2.4 UTR Related SNPs 3 Commonly Used Computational Intelligence Methods 3.1 Support Vector Machine 3.2 Artificial Neural Network 3.3 Decision Trees 3.4 Bayesian Networks 4 Predictive Models in Cancer Diagnosis 4.1 Multiple Myeloma 4.2 Breast Cancer 4.3 Lung Cancer 5 Challenges and Future Outlook 6 Conclusion References Realizing the Potential of Blockchain in Cancer Research 1 Introduction 2 Need for Blockchain Technology 3 Features of Blockchain 3.1 Decentralized 3.2 Distributed 3.3 Immutable 3.4 Transparent 3.5 Peer-To-Peer (P2P) Network 4 How Blockchain Works? 5 Convergence of Technology and Health Care 6 Drawbacks of Traditional Health Care in Cancer Research 7 Applications of Blockchain Technology in Cancer Research 7.1 Medical Record Management 7.2 Palliative Care 7.3 Future Research 7.4 Drugs and Equipment Regulation 7.5 Insurance and Cost-Saving 8 Future Directions 9 Conclusion References