Text Mining Approaches for Biomedical Data

The book 'Text Mining Approaches for Biomedical Data' delves into the fascinating realm of text mining in healthcare. It provides an in-depth understanding of how Artificial Intelligence (AI) and Machine Learning (ML) are revolutionizing healthcare research and patient care. The book covers a wide range of topics such as mining textual data in biomedical and health databases, analyzing literature and clinical trials, and demonstrating various applications of text mining in healthcare. This book is a guide for effectively representing textual data using vectors, knowledge graphs, and other advanced techniques. It covers various text mining applications, building descriptive and predictive models, and evaluating them. Additionally, it includes building machine learning models using textual data, covering statistical and deep learning approaches. This book is designed to be a valuable reference for computer science professionals, researchers in the biomedical field, and clinicians. It provides practical guidance and promotes collaboration between different disciplines. Therefore, it is a must-read for anyone who is interested in the intersection of text mining and healthcare. Preface Acknowledgements Contents Editors and Contributors 1 Introduction 1.1 Why We Write This Book? 1.2 Why Text Mining? 1.3 Why Biomedical and Text Mining? 1.4 What Are the Contents of This Book and Why? 1.5 Overview of Book Contents 1.6 Scope, Challenges and Future Directions? References Part I Domain Knowledge in Text Mining and Biomedical Data 2 Biomedical Data Types, Sources, Content, and Retrieval 2.1 What Is Biomedical Data? 2.2 Data Types and Storage 2.3 Data Sources and Formats 2.3.1 Sequence Data 2.3.2 Structural Data 2.3.3 Chemical Structure Data 2.3.4 Spectral Data 2.3.5 Image Data 2.3.6 Textual Data 2.4 Database Searching and Retrieval 2.5 Summary References 3 Information Analysis Using Biomedical Text Mining 3.1 Introduction 3.2 Information Retrieval 3.3 Information Extraction 3.4 Knowledge Discovery 3.5 Hypothesis Formation 3.6 Summary References 4 Connection and Curation of Corpus (Labeled and Unlabeled) 4.1 Introduction 4.2 Biomedical Literature and Data Types 4.3 Data Assimilations 4.4 Data Integration 4.5 Accessing Biomedical Literature 4.6 Data Curation 4.6.1 Biocuration 4.6.2 Methods of Curation 4.6.3 Steps for Biomedical Data Curation 4.6.4 Importance of Data Curation in Biomedical Domain 4.7 Data Curation Tools 4.8 Curated Biomedical Datasets 4.9 Cases Study—Adverse Drug Reaction Data 4.9.1 Dataset Description 4.9.2 Entities of Interest 4.9.3 Construction Steps 4.9.4 Annotated Corpus 4.10 Challenges in Biomedical Data Curation 4.11 Summary References 5 Biomedical Data Visualization 5.1 Introduction 5.2 Data Visualization and Human Body 5.3 Biomedical Data Visualization: Principles 5.4 Biomedical Data Visualization: Challenges 5.5 Perceptron Factors 5.6 Virtual Reality, Augmented Reality, and Mixed Reality 5.7 Current Challenges of VR/AR/MR Platforms 5.8 Biomedical Data Visualization Tools 5.9 Summary References 6 Biomedical Text Data Visualization 6.1 Biomedical Text Data: An Overview 6.2 Biomedical Text Data Visualization Using Word-Clouds 6.2.1 Word-Clouds—A Type of Visualization 6.3 Text Data Visualization Through Embeddings 6.3.1 t-SNE—A Type of Visualization 6.3.2 Word2Vec and Sent2Vec of Biomedical Data 6.4 Summary References Part II Biomedical Ontology and Model Building 7 Role of Ontology in Biomedical Text Mining 7.1 Introduction 7.2 Using Ontologies 7.3 Biomedical Text Mining 7.4 Using Ontologies in TM Techniques 7.5 Summary References 8 Ontology in Text Mining and Matching 8.1 Introduction 8.2 Types of Ontology 8.3 Semantic Web 8.4 Natural Language Processing 8.5 Ontology Matching 8.5.1 Similarity Measures 8.5.2 Ontology Matching Process 8.5.3 Evaluation Framework 8.5.4 Machine Learning for Biomedical Ontology Matching 8.5.5 Challenges in Ontology Matching 8.6 Data Quality/Data Variations 8.7 Datasets for Ontology Matching 8.8 The Future: Questions and Challenges 8.9 Summary References: 9 Fundamentals of Vector-Based Text Representation and Word Embeddings 9.1 Introduction 9.2 Bag of Word-Based Approach for Text Representation: An Overview 9.2.1 Preprocessing and Indexing for Constructing Bag of Words 9.2.2 Inverted Indexing 9.3 Classical Representation Models 9.3.1 Boolean Model 9.3.2 Vector Space Model (VSM) 9.4 Language Models and Word Embeddings 9.4.1 Language Models: An Overview 9.4.2 Word Embeddings 9.5 Summary References 10 Transformer-Based Models for Text Representation and Processing 10.1 Introduction 10.2 Need for Transformer-Based Models 10.3 Architecture of a Transformer-Based Model 10.4 BERT 10.4.1 Masked LM (MLM) 10.4.2 Next Sentence Prediction (NSP) 10.5 BioBERT 10.5.1 Pre-training BioBERT 10.5.2 Fine-Tuning BioBERT 10.6 Research Challenges in Text Representation 10.7 Summary References Part III Tasks in Biomedical Text Mining 11 Information Retrieval and Query Expansion for Biomedical Data 11.1 Introduction to Information Retrieval 11.2 Vector Space Model (VSM) for Information Retrieval 11.2.1 Classical Vector Space Model 11.2.2 Advanced VSM Based on Word Embedding 11.2.3 Indexing and Matching in VSM 11.3 Evaluation of Information Retrieval 11.4 Information Retrieval in Biomedical Domain 11.4.1 Benchmark IR Datasets in Biomedical IR 11.4.2 CLEF eHealth Lab Series 11.5 Need for Query Expansion (QE) in IR 11.6 Query Expansion: An Overview 11.7 Types of Query Expansion 11.8 Approaches for Query Expansion 11.8.1 Global Analysis 11.8.2 Local Analysis 11.9 Summary Annexure 1 Annexure 2 References 12 Advances in Biomedical Entity and Relation Extraction: Techniques and Applications 12.1 Introduction 12.2 Entity Extraction from Biomedical Text: An Overview 12.3 Approaches for Named Entity Recognition 12.4 Tools for NER 12.4.1 Metamap 12.4.2 Facta+ 12.4.3 Biomedical Entity Search Tool (BEST) 12.4.4 Hunflair 12.4.5 BERN2 12.4.6 Bio-epidemiology-NER 12.4.7 Output: 12.4.8 SciSpacy 12.5 Named Entity Normalization 12.6 Relation Extraction 12.6.1 Relation Extraction Using Co-occurrence 12.6.2 Pattern-Based Relation Extraction 12.6.3 Rule-Based 12.6.4 Machine Learning-Based Relation Extraction 12.7 Tools for Relation Extraction 12.7.1 Psychiatric Disorders Gene Association NETwork (PsyGeNET) 12.7.2 DisGeNet 12.7.3 The SIGnaling Network Open Resource (Signor3) 12.8 Applications 12.9 Challenges in Biomedical Entity Extraction 12.10 Summary References 13 Deep Learning for Extracting Biomedical Entities from COVID-19 Dataset: A Case Study 13.1 Introduction 13.2 CORD-NER Dataset 13.3 Deep Learning Architectures for Named Entity Recognition 13.4 Experiments on CORD-NER Using Deep Learning 13.4.1 Experiment on CORD-NER Using LSTM 13.4.2 Experiment on CORD-NER Using BioBERT 13.5 Summary References 14 Multilabel Text Classification in Biomedical Domain 14.1 Introduction 14.2 Types of Classification 14.3 Multilabel Classification 14.3.1 Benchmark Datasets for Biomedical Multilabel Classification 14.3.2 Evaluation Metrics for Multilabel Classification 14.4 Feature Extraction and SVM-Based Model for Biomedical Text Classification 14.4.1 Feature Extraction for Biomedical Text Classification 14.4.2 Experiments Using SVM for Biomedical Text Classification 14.5 Issues and Challenges for Multilabel Text Classification 14.6 Deep Learning Models for Text Classification 14.6.1 Emergence of Transformer-Based Deep Learning Models 14.6.2 Domain-Specific Vectors 14.6.3 Framework of BioBERT for Multilabel Classification 14.7 Summary References 15 Biomedical Document Clustering 15.1 Introduction 15.2 Word/Document Representation 15.2.1 TF-IDF-Based Representation 15.2.2 BERT-Based Representation 15.3 Similarity Measures 15.3.1 Vector-Based Similarity Measure 15.3.2 Ontology-Based Text Similarity Measures 15.4 Clustering Algorithm 15.4.1 K-means Clustering 15.5 Experiment and Result 15.5.1 Dataset 15.5.2 Experiments 15.5.3 Result and Visualization References Part IV Knowledge Graph for Biomedical Text Mining 16 Exploring Knowledge Graphs (KG): A Comprehensive Overview 16.1 Introduction 16.2 Definition and Representation of Knowledge Graph 16.3 Why Are KGs Required in Biomedical Text Mining? 16.4 Existing Knowledge Graphs in Biomedical Domain 16.4.1 A Knowledge Graph for Ebola Virus (EBOV) 16.4.2 A Knowledge Graph Constructed from Electronic Medical Records (EMRs) 16.4.3 A Question–Answer Derived Knowledge Graph for Accelerating Clinical Research 16.4.4 QIAGEN Knowledge Base (KB) Enables the Construction of Various Biomedical Knowledge Graphs 16.4.5 A Knowledge Graph for Understanding Host-Cell Interaction upon COVID-19 Infection 16.4.6 Chem2Bio2RDF 16.5 Benefits and Limitations of KGs 16.6 Applications of KG in the Biomedical Domain 16.7 Summary References 17 Building Knowledge Graphs in the Biomedical Domain: Methods and Case Studies 17.1 Introduction 17.2 Data Sources for the Construction of Biomedical Knowledge Graphs 17.3 Construction of Knowledge Graphs 17.4 Case Study: Automatic Knowledge Graph Construction from Biomedical Abstracts 17.5 Demo 17.6 Summary References Part V Applications of Biomedical Text Mining 18 Text Mining for Telemedicine 18.1 Introduction to Telemedicine 18.2 Types of Telemedicine Services 18.3 Telemedicine: Tools and Systems 18.4 Usage of Text Mining for Telemedicine Services 18.5 Challenges in Telemedicine 18.6 Summary References 19 Text Mining for Recommendation Systems/Expert Systems in Health Domain 19.1 What Is a Recommendation System? 19.2 Types of Recommendation Systems 19.2.1 Content-Based Filtering (CB) 19.2.2 Collaborative Filtering (CF) 19.2.3 Knowledge-Based Recommender System (KB) 19.2.4 Hybrid Recommender Systems (HyR) 19.3 Basic Architecture of Healthcare Recommendation Systems 19.4 Available Tools for Building RS 19.5 Challenges 19.6 Summary References 20 Agent-Based Modeling and Simulation, with Emphasis on Healthcare Data 20.1 Introduction 20.1.1 What Is an Agent? 20.1.2 What Is the Need for ABMS? 20.1.3 When Is ABMS Useful? 20.2 Tools for ABMS 20.2.1 Large-Scale Agent Development Environments 20.3 Application of ABMS in Healthcare 20.3.1 Infectious Disease Epidemiology 20.3.2 Non-communicable Disease Control 20.3.3 Simulating Health Behavior 20.3.4 Hospital MRSA Infection Management 20.3.5 Tissue Remodeling and Wound Healing 20.3.6 BioWar 20.4 Advantages and Limitations of ABMS 20.5 Challenges and Opportunities in ABMS 20.6 A Way Forward References 21 Ethical Issues in Biomedical Text Mining 21.1 Introduction 21.1.1 Importance and Challenges of Ethical Considerations in Biomedical Text Mining 21.2 Ethics 21.2.1 Ethical Concerns at the Academia/Industry Level 21.2.2 Ethical Issues with Data 21.3 Solutions for Ethical Issues with Data 21.4 Bias Cases 21.5 Regulations 21.6 Need for Standardization 21.7 Different Treatments 21.8 Recommended Practice 21.9 Summary References