Advances in Artificial Intelligence, Computation, and Data Science: For Medicine and Life Science

Preface
Contents
Part I Bioinformatics
1 Intelligent Learning and Verification of Biological Networks
1.1 Introduction
1.2 Statistical Learning of Regulatory Networks
1.2.1 INSPECT Change-Points Identification
1.2.2 Network Structure Learning and Searching
1.2.3 Regulatory Relationship Identification
1.3 Formal Analysis of Regulatory Networks
1.3.1 Temporal Logic Formula
1.3.2 Symbolic Model Checking
1.3.3 Time-Bounded Linear Temporal Logic (BLTL)
1.3.4 Probabilistic Model Checker PRISM
1.4 Integrative Data Analysis
1.5 Discussions
References
2 Differential Expression Analysis of RNA-Seq Data and Co-expression Networks
2.1 Systems Biology
2.2 High Throughput Sequencing
2.3 RNA-seq Analysis
2.4 Formulating a Sequencing Library
2.5 Biological and Technical Variations
2.6 Assessment of Variations
2.6.1 Poisson’s Distribution
2.6.2 Negative Binomial Distribution
2.7 Method for Differential Expression Analysis
2.8 Generalized Linear Model (GLM)
2.9 Hypothesis Test
2.10 Normalization of Data
2.11 Trimmed Mean of M-values (TMM)
2.12 Relative Log Expression (RLE)
2.13 Upper-Quartile Normalization
2.14 Principal Component Analysis
2.14.1 Steps of PCA Analysis
2.15 Data Analysis of Gene Expression Profiles
2.16 An Illustration: A Differential Gene Expression Analysis Conducted on a Real Dataset
2.17 R Packages Used in the RNA-Seq Analysis
2.18 Removal of Lowly Transcribed Genes
2.19 Formation of DGEList Object Using EdgeR
2.20 Density Distributions
2.21 Normalization
2.22 Principal Component Analysis
2.23 Design Matrix
2.24 NB and QL Dispersion Evaluation
2.25 Annotating Genes
2.26 Gene Testing
2.27 GO Analysis
2.28 ROAST Analysis
2.29 CAMERA Test
2.30 Visualizing Gene Tests
2.31 Graph Theory Terminologies
2.32 Gene Regulatory Network (GRN)
2.33 Inference of Gene Regulatory Networks
2.34 Gene Regulatory Network Modelling
2.35 Correlation and Partial Correlation-based Methods
2.36 Co-expression Networks
2.37 Pre-processing of Data
2.38 Construction of Covariance Matrix
2.39 Measure of Similarity
2.40 Network Construction
2.41 Module Detection
2.42 Module Enrichment
2.43 WGCNA Package in R
2.44 Co-expression Network Analysis with Real Dataset
2.45 Concluding Remarks
References
3 Learning Biomedical Networks: Toward Data-Informed Clinical Decision and Therapy
3.1 Biological Data and the Rise of Targeted Therapies
3.2 Network Analysis in Biomedical Informatics
3.2.1 Differential Network Analysis
3.2.2 Network-Based Regularization
3.2.3 Causal Discovery and Inference
3.3 Software and Biomedical Applications
3.4 Conclusions and Future Work
References
4 Simultaneous Clustering of Multiple Gene Expression Datasets for Pattern Discovery
4.1 Simultaneous Clustering Methods
4.1.1 Cluster of Clusters (COCA)
4.1.2 Bi-CoPaM
4.1.3 UNCLES and M–N Scatter Plots
4.1.4 Clust
4.1.5 Deep Learning Approaches
4.2 Case Study 1: A Novel Subset of Genes with Expression Consistently Oppositely Correlated with Ribosome Biogenesis in Forty Yeast Datasets
4.2.1 Data and Approach
4.2.2 Results and Discussion
4.2.3 Summary and Conclusions
4.3 Case Study 2: A Transcriptomic Signature Derived from a Study of Sixteen Breast Cancer Cell-Line Datasets Predicts Poor Prognosis
4.3.1 Data and Approach
4.3.2 Results and Discussion
4.3.3 Summary and Conclusions
4.4 Case Study 3: Cross-Species Application of Clust Reveals Clusters with Contrasting Profiles Under Thermal Stress in Two Rotifer Animal Species
4.5 Summary and Conclusions
References
5 Artificial Intelligence for Drug Development
5.1 Introduction
5.2 Methodologies in Pre-clinical and Clinical Trials
5.3 Post-Market Trials
5.4 Concluding Remarks
References
6 Mathematical Bases for 2D Insect Trap Counts Modelling
6.1 Introduction
6.2 Mean Field and Mechanistic Models of Insect Movement with Trapping
6.2.1 Isotropic Diffusion Model and Computing Trap Counts
6.2.2 Individual Based Modelling Using Random Walks
6.2.3 Simple Random Walk (SRW)
6.2.4 Simulating Trapping
6.2.5 Equivalent Trap Counts
6.3 Geometrical Considerations for Trap Counts Modelling
6.3.1 Simulation Artefacts Due to the RW Jump Process
6.3.2 Impact of the Arena Boundary Shape, Size and the Average Release Distance
6.3.3 Impact of Trap Shape
6.4 Anisotropic Models of Insect Movement
6.4.1 Correlated Random Walk (CRW)
6.4.2 MSD Formula for the CRW
6.4.3 Measuring Tortuosity
6.4.4 Biased Random Walk (BRW)
6.4.5 MSD Formula for the BRW
6.4.6 Equivalent RWs in Terms of Diffusion
6.4.7 Drift Diffusion Equation
6.4.8 Biased and Correlated Random Walk (BCRW)
6.5 Effect of Movement on Trap Counts
6.5.1 Effect of Movement Diffusion
6.5.2 Baited Trapping
6.6 Concluding Remarks
References
Part II Medical Image Analysis
7 Artificial Intelligence in Dermatology: A Case Study for Facial Skin Diseases
7.1 Introduction
7.2 State of the Art
7.3 Study Case
7.3.1 Considered Skin Diseases
7.3.2 Machine-Learning/Deep-Learning Approaches
7.3.3 Preliminary Results
7.4 Developed Software
7.4.1 Patient Actions
7.4.2 Doctor Actions
7.5 Conclusion
References
8 Medical Imaging Based Diagnosis Through Machine Learning and Data Analysis
8.1 Introduction
8.2 Classification
8.2.1 Classifiers
8.2.2 Example 1: Similarity Metric
8.2.3 Example 2: Similarity Learning
8.3 Dense Prediction
8.3.1 Segmentation
8.3.2 Synthesis
8.4 Multi-modality Analysis
8.4.1 Example: A Non-deep-Learning Based Approach for Multi-modal Feature Selection
8.4.2 Example: A Deep Learning Based Approach for Multi-modality Fusion
8.5 Conclusion
References
9 EfficientNet-Based Convolutional Neural Networks for Tuberculosis Classification
9.1 Introduction
9.2 Related Work
9.3 Methodology
9.4 Description of Dataset
9.5 Results and Discussions
9.5.1 Feature Visualization of pretrained models for TB classification
9.6 Conclusions
References
10 AI in the Detection and Analysis of Colorectal Lesions Using Colonoscopy
10.1 Introduction
10.1.1 Colorectum and Colorectal Cancer
10.1.2 Colorectal Cancer Stages
10.1.3 Colonoscopy and Colorectal Polyps
10.1.4 Application of AI in Colonoscopy
10.2 Computer-Aided Detection in Colorectal Polyps
10.2.1 Why Computer-Aided Detection
10.2.2 Early Computer-Aided Detection Algorithm
10.2.3 Recent Computer-Aided Detection Algorithms
10.3 Computer-Aided Classification in Colorectal Polyps
10.3.1 Why Computer-Aided Classification
10.3.2 Early Computer-Aided Analysis (CADx)
10.3.3 Recent Progress of CADx
10.3.4 Limitations of CADx
10.4 Conclusion
References
11 Deep Learning-Driven Models for Endoscopic Image Analysis
11.1 Introduction
11.2 Deep Learning Architectures
11.2.1 Convolutional Neural Networks for Image Classification
11.2.2 Region-Level CNNs for Lesion Detection
11.2.3 Fully Convolutional Neural Networks for Segmentation
11.3 Case Study I: Gastrointestinal Hemorrhage Recognition in WCE Images
11.3.1 Background of the Application
11.3.2 Improved Learning Strategy
11.3.3 Dataset
11.3.4 Evaluation Metrics
11.3.5 Experimental Results
11.4 Case Study II: Colorectal Polyp Recognition in Colonoscopy Images
11.4.1 Background of the Application
11.4.2 Improved Learning Strategy
11.4.3 Dataset
11.4.4 Evaluation Metrics
11.4.5 Experimental Results
11.5 Conclusion and Future Perspectives
References
Part III Physiology
12 A Dynamic Evaluation Mechanism of Human Upper Limb Muscle Forces
12.1 Introduction
12.2 Related Work
12.3 Materials and Methods
12.3.1 Data Collection and Preprocessing
12.3.2 Joint Angle Estimation
12.3.3 OpenSim Simulation
12.3.4 Muscle Activation Dynamics
12.4 Results
12.5 Discussion
12.6 Conclusions
References
13 Resting-State EEG Sex Classification Using Selected Brain Connectivity Representation
13.1 Introduction
13.2 Related Work
13.3 Data and Methods
13.3.1 Dataset Description
13.3.2 Preprocessing
13.3.3 Signal Representation
13.3.4 Feature Analysis
13.3.5 Classification
13.4 Results
13.4.1 Feature Selection
13.4.2 Validation Results
13.4.3 Test Results
13.5 Conclusions
References
Part IV Innovation in Medicine and Health
14 Augmented Medicine: Changing Clinical Practice with Artificial Intelligence
14.1 Introduction
14.2 Implementation of Augmented Medicine in Clinical Practice: An Overview
14.2.1 Monitoring with Wearable Technology
14.2.2 AI for Diagnosis
14.2.3 Machine Learning for Prediction
14.3 Conclusions
References
15 Environmental Assessment Based on Health Information Using Artificial Intelligence
15.1 Introduction
15.2 Environmental Parameters and Health
15.2.1 Air Pollution
15.2.2 Weather-Related Parameters
15.2.3 Illumination
15.2.4 Implications for Health-Related BACS
15.3 System Concept for Health based Environmental Assessment
15.3.1 System Components and Their Interactions
15.3.2 Data Interpretation for Medical Staff
15.3.3 Feedback Systems for Patients
15.3.4 Communication Between BACS and EHR
15.4 Approaches for Environmental Assessment
15.4.1 Data Organization
15.4.2 Derived Constraints
15.4.3 Appropriate Methods for Environmental Risk Estimation
15.5 Conclusion and Discussion
References
Index