Soft Computing: Biomedical and Related Applications (Studies in Computational Intelligence, 981)

✍ Scribed by Nguyen Hoang Phuong (editor), Vladik Kreinovich (editor)

• Publisher: Springer
• Year: 2021
• Tongue: English
• Leaves: 319
• Category: Library

✦ Synopsis

This book lists current and potential biomedical uses of computational intelligence methods. These methods are used in diagnostics and treatment of such diseases as cancer, cardiac diseases, pneumonia, stroke, and COVID-19. Many biomedical problems are difficult; so, often, the current methods are not sufficient, new methods need to be developed. To confidently apply the new methods to critical life-and-death medical situations, it is important to first test these methods on less critical applications. The book describes several such promising new methods that have been tested on problems from agriculture, computer networks, economics and business, pavement engineering, politics, quantum computing, robotics, etc.

This book helps practitioners and researchers to learn more about computational intelligence methods and their biomedical applications―and to further develop this important research direction.

✦ Table of Contents

Preface
Contents
Biomedical Applications of Computational Intelligence Techniques
Bilattice CADIAG-II: Theory and Experimental Results
1 Introduction
1.1 Background
1.2 CADIAG Systems
1.3 Objective
2 Background—CADIAG-II
2.1 Overall Consultation Process
2.2 Knowledge Representation
3 Bilattice CADIAG-II
3.1 Algebraic Preliminaries on Bilattices
3.2 An Extension of CADIAG-II Based on Bilattices—bCADIAG-II
4 Implementation and Experimental Results
4.1 MedFrame, CADIAG-II, and bCADIAG-II
4.2 Evaluation and Results
4.3 Discussion of Results
5 Conclusions
References
A Combination Model of Robust Principal Component Analysis and Multiple Kernel Learning for Cancer Patient Stratification
1 Introduction
2 Method
2.1 Robust Principal Component Analysis
2.2 Cancer Patient Stratification Model
2.3 Dimensionality Reduction and Features Extraction Based on RPCA
2.4 Cancer Patient Stratification Model Based on Multiple Kernel Learning
3 Materials and Experiment
3.1 Materials
3.2 Experiment
4 Results and Discussion
5 Conclusion
References
Attention U-Net with Active Contour Based Hybrid Loss for Brain Tumor Segmentation
1 Introduction
2 Materials and Methods
3 The Proposed Approach
3.1 The Pipeline of the Proposed Approach
3.2 Loss Function
3.3 Data Augmentation and Training
4 Evaluations and Results
4.1 Dataset
4.2 Results
4.3 Evaluation on other Networks
4.4 Performance of the Proposed Loss Function
5 Conclusion
References
Refining Skip Connections by Fusing Multi-scaled Context in Neural Network for Cardiac MR Image Segmentation
1 Introduction
2 Materials and Methods
2.1 Network Architecture
2.2 The Proposed Skip Module
2.3 The Proposed Attention Box
2.4 Performance Metrics
2.5 Training
3 Results
4 Compare to Other Methods
5 Conclusion
References
End-to-End Hand Rehabilitation System with Single-Shot Gesture Classification for Stroke Patients
1 Introduction
2 Methodology
2.1 MediaPipe
2.2 Derivation of the Degrees of Freedom of Each Finger's Joint
2.3 Gesture Recognition
2.4 Gamification
3 Implementation and Results
4 Discussion and Conclusion
References
Feature Selection Based on Shapley Additive Explanations on Metagenomic Data for Colorectal Cancer Diagnosis
1 Introduction
2 Related Work
3 Datasets for Colorectal Cancer Prediction Tasks
4 Learning Model and Feature Selection Methods-Based Pearson and SHAP
4.1 Learning Model
4.2 Feature Selection Approach
5 Experimental Results
6 Conclusion
References
Clinical Decision Support Systems for Pneumonia Diagnosis Using Gradient-Weighted Class Activation Mapping and Convolutional Neural Networks
1 Introduction
2 Related Work
3 Chest X-Ray Images for Pneumonia Classification Experiments
4 Method
4.1 Data Augmentation and Transfer Learning
4.2 The Proposed Convolutional Neural Network Architecture
4.3 Explainable Deep Learning Using Grad-CAM
5 Results
5.1 Model Evaluation
5.2 Experimental Results
5.3 Comparative of Explanations of the Results Using Grad-CAM
5.4 Performance Analysis
5.5 The Pneumonia Diagnosis Based on Chest X-Ray System
6 Conclusion
References
Improving 3D Hand Pose Estimation with Synthetic RGB Image Enhancement Using RetinexNet and Dehazing
1 Introduction
2 Approach
2.1 Low Light Enhancement (RetinexNet)
2.2 Dark Channel Prior Dehazing
3 Experimental Results and Discussion
4 Conclusions and Future Work
References
Imbalance in Learning Chest X-Ray Images for COVID-19 Detection
1 Introduction
2 Related Work
3 The Method
3.1 Feature by Deep Learning
3.2 Feature Refining by Checking Class Imbalance
3.3 Algorithm
4 Experiments
5 Conclusions
References
Deep Learning Based COVID-19 Diagnosis by Joint Classification and Segmentation
1 Introduction
2 Methodology
2.1 Dataset Description
2.2 Proposed Method
2.3 Segmentation Stage
2.4 Classification Stage
3 Training Protocol
4 Experimental Results
5 Conclusion
References
General Computational Intelligence Techniques and Their Applications
Why It Is Sufficient to Have Real-Valued Amplitudes in Quantum Computing
1 Formulation of the Problem
2 Our Explanation
References
On an Application of Lattice-Valued Integral Transform to Multicriteria Decision Making
1 Introduction
2 Preliminaries
2.1 Algebra of Truth Values
2.2 Fuzzy Sets
2.3 Fuzzy Measure Spaces
2.4 Multiplication-Based Fuzzy Integral
2.5 Integral Transform
3 MCDM Based on the Integral Transform
4 Illustrative Example
5 Conclusion
References
Fine-Grained Network Traffic Classification Using Machine Learning: Evaluation and Comparison
1 Introduction
2 Related Work
2.1 Fine-Grained Network Traffic Classification
2.2 Machine Learning Algorithms
2.3 Evaluation Metrics
3 Method
4 Experiments
4.1 Datasets
4.2 Data Preprocessing
4.3 Training and Evaluation of Machine Learning Models
5 Conclusions
References
Soil Moisture Monitoring System Based on LoRa Network to Support Agricultural Cultivation in Drought Season
1 Introduction
2 Proposed System Architecture
2.1 Sensor Node
2.2 Gateway
3 Experiment Results
3.1 EC-5 Soil Sensor Laboratory Calibration Experiment
3.2 Network Coverage Test
3.3 Field Test
4 Discussion and Conclusion
References
Optimization Under Fuzzy Constraints: Need to Go Beyond Bellman-Zadeh Approach and How It Is Related to Skewed Distributions
1 Formulation of the Problem
2 Main Idea and the Resulting Definition
References
Towards Parallel NSGA-II: An Island-Based Approach Using Fitness Redistribution Strategy
1 Introduction
2 Preliminaries
2.1 Genetics Algorithm (GA)
2.2 Non-dominated Sorting Genetic Algorithm (NSGA)
2.3 NSGA-II
3 Related Work
4 Underlying Ideas
5 The Model-Based Approach to Parallel NSGA-II with Fitness Redistribution
5.1 Sub-populations' Organization and Evolution Process
5.2 Fitness Redistribution Strategy
5.3 Race Condition Problem in Parallel Algorithm
6 Experiments
6.1 Experiments with Deterministic Problems
6.2 Experiments with Non-deterministic Problems
7 Conclusion
References
A Radial Basis Neural Network Approximation with Extended Precision for Solving Partial Differential Equations
1 Introduction
2 Proposed Numerical Procedure
2.1 IRBFN High-Order Approximations
2.2 Flat Kernel Integrated RBF
3 Numerical Examples
3.1 Flat IRBF Solution of ODE
3.2 Flat IRBF Solution of PDE
4 Conclusion
References
Why Some Power Laws Are Possible and Some Are Not
1 Formulation of the Problem
2 Power Laws and Scale Invariance: A Brief Reminder
3 Main Idea and Resulting Explanation
References
How to Estimate the Stiffness of a Multi-layer Road Based on Properties of Layers: Symmetry-Based Explanation for Odemark's Equation
1 Formulation of the Problem
2 Scale-Invariance: Reminder
3 Towards an Explanation
4 Proof of the Main Result
References
Need for Diversity in Elected Decision-Making Bodies: Economics-Related Analysis
1 Diversity and Economics-Related Decision Making: Formulation of the Problem
2 Accurate Economics-Related Decision Making Model and How Its Optimization Implies Diversity
References
Fuzzy Transform for Fuzzy Fredholm Integral Equation
1 Introduction
1.1 Fuzzy Numbers and Fuzzy Arithmetic Operations
1.2 Fuzzy Fredholm Integral Equations
1.3 F-Transform of Functions of Two Variables
2 Function Approximation
2.1 Some Auxiliary Properties of and ψ
3 General Scheme of the Proposed Method
4 Existence of a Unique Solution
4.1 Existence of a Unique Solution to Eq. (1)
4.2 Existence of a Fuzzy Approximate Solution
5 Conclusion
References
Constructing an Intelligent Navigation System for Autonomous Mobile Robot Based on Deep Reinforcement Learning
1 Introduction
2 Kinematic Model of AMR Robot
3 Design a Intelligent Navigation System Based on Deep Reinforcement Learning
3.1 Input Normalization
3.2 Construct a Network Structure
4 Simulation
4.1 Simulation Result
4.2 Experiment Result
5 Conclusion
References
One-Class Support Vector Machine and LDA Topic Model Integration—Evidence for AI Patents
1 Introduction
2 Related Literature
3 Data
4 Method
4.1 One-Class Support Vector Machines
4.2 Topic Modeling
4.3 Training Class
5 Results
6 Conclusion
References
HDBSCAN: Evaluating the Performance of Hierarchical Clustering for Big Data
1 Introduction
2 Related Works
3 Efficient Spark HDBSCAN
3.1 Choose the Configuration
3.2 Separation Problem
3.3 The Parallel HDBSCAN Algorithms
4 Experimental Evaluation
4.1 Experimental Setup
4.2 Datasets
4.3 Runtime Evaluation
4.4 The Merge Factor K
4.5 The Speedup Score
5 Conclusions and Future Works
References
Applying Deep Reinforcement Learning in Automated Stock Trading
1 Introduction
2 The Simulated World of Stock Markets
2.1 Stock Market as an MDP Framework
2.2 Agent in the Simulated World
2.3 Optimization Goals in the Simulated World
3 A Deep Reinforcement Learning Trading Method
3.1 A Brief Summary of Reinforcement Learning
3.2 Motivation for Applying Deep Reinforcement Learning
3.3 Actor-Critic Network
3.4 Twin Delayed Deep Deterministic Policy Gradient (TD3)
3.5 Soft Actor-Critic (SAC)
4 Experimental Setup and Evaluation Method
4.1 Data Preparation and Necessary Toolkits
4.2 Experimental Setup and Walk-Forward Cross-validation
4.3 Results
5 Conclusions
References
Telecommunications Services Revenue Forecast Using Neural Networks
1 Introduction
2 Related Work
3 The Proposed Method for Revenue Forecast
3.1 Input Determination
3.2 Data Collection
3.3 Determining the Forecast Period
3.4 Data Preprocessing
3.5 Training and Test Sets Division for Model Learning and Evaluation
4 Experimental Results
4.1 Metrics for Evaluation
4.2 Neural Networks Setting for Forecasting
4.3 Revenue Forecasting
5 Conclusion
References
Product Recommendation System Using Opinion Mining on Vietnamese Reviews
1 Introduction
2 Related Work
3 Recommendation System Based on Functions/Features of the Product
3.1 Data Collection
3.2 Module Building Themes for Feature Sets
4 Experimental Results
4.1 Data Collection and Database for Recommender System
4.2 Scenarios
4.3 Discussion
5 Conclusion
References