Nature-Inspired Optimization Methodologies in Biomedical and Healthcare

Description

This book introduces a variety of well-proven and newly developed nature-inspired optimization algorithms solving a wide range of real-life biomedical and healthcare problems. Few solo and hybrid approaches are demonstrated in a lucid manner for the effective integration and finding solution for a large-scale complex healthcare problem. In the present bigdata-based computing scenario, nature-inspired optimization techniques present adaptive mechanisms that permit the understanding of complex data and altering environments. This book is a voluminous collection for the confront faced by the healthcare institutions and hospitals for practical analysis, storage, and data analysis. It explores the distinct nature-inspired optimization-based approaches that are able to handle more accurate outcomes for the current biomedical and healthcare problems. In addition to providing a state-of-the-art and advanced intelligent methods, it also enlightens an insight for solving diversified healthcare problems such as cancer and diabetes. Foreword Preface Contents 1 Nature-Inspired Optimization Algorithms: Past to Present 1.1 Introduction 1.1.1 Why Do We Need Nature-Inspired Optimization Algorithms? 1.1.2 Classification of Optimization Algorithms 1.2 Background 1.2.1 Natural Computing 1.2.2 Algorithm 1.2.3 Optimization 1.2.4 Metaheuristic 1.3 Broad Review on Nature-Inspired Optimization Algorithms 1.3.1 Genetic Algorithms 1.3.2 Ant Colony Optimization 1.3.3 Swarm Intelligence 1.3.4 Artificial Bee Colony (ABC) 1.3.5 ACO-Ant Colony Optimization 1.3.6 BAT Algorithm 1.4 Theoretical Analysis and Applications 1.4.1 Applications 1.5 Discussions, Challenges, Open Issues and Future Recommendations 1.6 Conclusion References 2 Preventing the Early Spread of Infectious Diseases Using Particle Swarm Optimization 2.1 Introduction 2.2 Literature Review of the Status of Research and Development in the Subject 2.3 Methodology 2.3.1 Pattern Prediction with Prior Knowledge 2.4 Experiments and Results 2.4.1 Running Environment 2.4.2 Performance Metric 2.5 Conclusion and Future Enhancement References 3 Optimized Gradient Boosting Tree-Based Model for Obesity Level Prediction from patient’s Physical Condition and Eating Habits 3.1 Introduction 3.2 Literature Study 3.3 Understanding Factors Associated with Obesity 3.4 Proposed Approach 3.4.1 Artificial Physics Optimization 3.4.2 APO Based GBT for Obesity Prediction 3.5 Experimental Result and Analysis 3.6 Conclusion References 4 Multi-Objective Optimization Algorithms in Medical Image Analysis 4.1 Introduction 4.2 Perceptual Method of Color Correction Based on Multi-Objective Optimization 4.2.1 Loss-Function for Perceptual Color Correction 4.2.2 Multi-Objective Optimization 4.2.3 Color Correction Method 4.3 Experimental Results 4.4 Conclusion References 5 Heart Failure Detection from Clinical and Lifestyle Information using Optimized XGBoost with Gravitational Search Algorithm 5.1 Introduction 5.2 Literature Survey 5.3 Exploratory Data Analysis of Heart Failure Data 5.4 Proposed Method 5.4.1 Gravitation Search Algorithm 5.4.2 GSA-Based XGB for Heart Failure Detection 5.5 Results and Analysis 5.6 Conclusion References 6 NIANN: Integration of ANN with Nature-Inspired Optimization Algorithms 6.1 Introduction 6.2 Literature Review 6.3 Integration of Artificial Neural Network and Optimization Algorithm 6.3.1 Artificial Neural Network (ANN) 6.3.2 Optimization Algorithm 6.3.3 Integration of ANN and OA 6.4 Experimental Results and Discussion 6.4.1 Experimental Setup 6.4.2 Discussion on ANN 6.5 Conclusion References 7 Hybridization of Fuzzy Theory and Nature-Inspired Optimization for Medical Report Summarization 7.1 Introduction 7.2 Literature Survey 7.3 Proposed Methodology 7.3.1 Preprocessing 7.3.2 Fuzzy C-Means Clustering 7.3.3 Defuzzification X-Cut 7.3.4 Generate the Base Summaries 7.3.5 Nature Inspired Optimization 7.4 Experimental Results 7.4.1 Experimental Setup 7.4.2 Performance Evaluation W.r.t ROUGE 7.4.3 Compare Performance with Different Summarising Approaches W.r.t ROUGE 7.5 Conclusion and Future Direction References 8 An Optimistic Bayesian Optimization Based Extreme Learning Machine for Polycystic Ovary Syndrome Diagnosis 8.1 Introduction 8.2 Related Work 8.3 Proposed Work 8.3.1 Extreme Learning Machine 8.3.2 Bayesian Optimization (BO) 8.3.3 Proposed ELM + BO Method 8.4 Discussion of Result Analysis and Simulation Setup 8.4.1 Dataset Overview and Environmental Setup 8.4.2 Result Analysis 8.5 Conclusion References 9 Diabetes Twitter Classification Using Hybrid GSA 9.1 Introduction 9.2 Related Works 9.3 Tweets Extraction 9.4 Methodology 9.4.1 GSA 9.4.2 CNN 9.4.3 GRU 9.4.4 LSTM 9.4.5 Embedding Layer 9.4.6 Dropout Layer 9.4.7 Maxpooling 9.4.8 Output Layer 9.5 Data Collection 9.6 Data Pre-processing 9.6.1 Conversion and Correction 9.6.2 Tokenization 9.6.3 Stop Words 9.6.4 Lemmatization 9.6.5 Word Stemming 9.6.6 Word Representation 9.6.7 Bag of Word 9.6.8 TF-IDF 9.6.9 Word2Vec 9.7 Proposed Capsule Network with GSA 9.7.1 Capsule Network 9.7.2 Capsule Network with GSA Algorithm 9.8 Conclusion References 10 Advance Machine Learning and Nature-Inspired Optimization in Heart Failure Clinical Records Dataset 10.1 Introduction 10.1.1 Cardiovascular Disease (CVD) 10.2 Related Works 10.3 Tree Based Algorithms 10.4 Natured Inspired Optimization (NIO) 10.5 Basic Preliminaries of Optimization Techniques 10.5.1 Bat Algorithm 10.5.2 Hybrid Bat Algorithm 10.5.3 Hybrid Self Adaptive Bat Algorithm 10.5.4 Firefly Algorithm 10.5.5 Grey Wolf Algorithm 10.6 Experiment Setup and Datasets Descriptions 10.6.1 System Environment 10.6.2 Heart Failure Clinical Records Data Set 10.7 Results 10.8 Conclusion and Future Work References 11 Early Detection of Chronic Obstructive Pulmonary Disease Using LSTM-Firefly Based Deep Learning Model 11.1 Introduction 11.2 Literature Study 11.3 Proposed Method 11.3.1 Firefly Optimization Algorithm 11.3.2 Long Short-Term Memory (LSTM) 11.3.3 LSTM + Firefly Methodology 11.4 Experimental Setup 11.4.1 Dataset 11.4.2 Simulation Environment 11.4.3 Performance Measures 11.5 Result Analysis 11.6 Critical Discussion 11.7 Conclusion References 12 GACO: A Genetic Algorithm with Ant Colony Optimization—Based Feature Selection for Breast Cancer Diagnosis 12.1 Introduction 12.2 Related Work 12.3 Preliminaries 12.3.1 Data Normalization 12.3.2 Principal Component Analysis (PCA) 12.3.3 Genetic Algorithm (GA) 12.3.4 Ant Colony Optimization (ACO) 12.3.5 Random Forest Algorithm (RF) 12.4 Proposed System 12.4.1 Data Preprocessing Component (DPC) 12.4.2 Evolutionary Algorithm-Based Feature Selection Component (EAFSC) 12.4.3 Proposed GACO_RF Component (GRC) 12.5 Results and Discussions 12.5.1 Dataset Description 12.5.2 Data Preprocessing 12.5.3 GA Parameters Estimation 12.5.4 GACO Parameter Estimation 12.5.5 Performance Comparison of Evolutionary Feature Selection Methods 12.5.6 Performance Metric 12.5.7 Performance Comparison of Proposed GACO_RF Model 12.6 Conclusion References