دانلود کتاب Statistical Methods at the Forefront of Biomedical Advances

فهرست مطالب :

Preface
Contents
1 Multivariate Disease Mapping Models to Uncover Hidden Relationships Between Different Cancer Sites
1.1 Introduction
1.2 M-Models for Multivariate Spatio-Temporal Areal Data
1.2.1 Model Implementation and Identifiability Constraints
1.3 Joint Analysis of Lung, Colorectal, Stomach, and LOCP Cancer Mortality Data in Spanish Provinces
1.3.1 Descriptive Analysis
1.3.2 Model Fitting Using INLA
1.4 Discussion
References
2 Machine Learning Applied to Omics Data
2.1 Introduction
2.2 Data Types
2.2.1 Genomics
2.2.2 Immunomics
2.3 Challenges in the Omics Data Analysis
2.4 Machine Learning Techniques
2.4.1 Random Forests
2.4.2 Multinomial Logistic Regression
2.4.3 Association Rules
2.5 Application
2.5.1 Study Subjects
2.5.2 Material and Methods
2.5.3 Results
2.5.3.1 Random Forest and LASSO Multinomial Logistic Regression
2.5.3.2 Association Rules
2.6 Conclusions and Future Work
Appendix
References
3 Multimodality Tests for Gene-Based Identification of Oncological Patients
3.1 Introduction
3.2 Analysing the Number of Groups
3.3 Application to the Gene Expression of Breast Cancer Patients
3.4 Conclusions and Future Work
Author Contribution Statement
Appendix
Genes Presenting a Multimodal Pattern
References
4 Hippocampus Shape Analysis via Skeletal Models and Kernel Smoothing
4.1 Introduction
4.2 Methodology
4.2.1 Kernel Smoothing on the Polysphere
4.2.1.1 Density Estimation
4.2.1.2 Gradient and Hessian Density Estimation
4.2.1.3 Polysphere-on-Scalar Regression Estimation
4.2.2 Density Ridges
4.2.2.1 Population Euclidean Case
4.2.2.2 Sample Polyspherical Case
4.2.2.3 Bandwidth Selection
4.2.2.4 Euler Iteration
4.2.2.5 Indexing Ridges
4.3 Results
4.3.1 An Illustrative Numerical Example
4.3.2 Main Mode of Variation of Hippocampus Shapes
4.4 Discussion
Proofs
References
5 Application of Quantile Regression Models for Biomedical Data
5.1 Introduction
5.2 The New Testing Procedure
5.2.1 Bootstrap Approximation
5.2.2 Computational Aspects
5.3 Simulation Study
5.4 Real Data Application
5.5 Conclusions
Appendix
References
6 Advances in Cytometry Gating Based on Statistical Distances and Dissimilarities
6.1 Introduction
6.2 Dissimilarities and Distances
6.2.1 Wasserstein Distance
6.2.2 Maximum Mean Discrepancy
6.2.3 Kullback–Leibler Divergence
6.2.4 Hellinger Distance
6.2.5 Friedman–Rafsky Statistic
6.3 Applications to the Gating Workflow
6.3.1 Grouping Cytometric Datasets
6.3.1.1 Ungated Cytometry Datasets
6.3.1.2 Gated Cytometry Datasets
6.3.2 Template Production
6.3.2.1 Ungated Cytometry Datasets
6.3.2.2 Gated Cytometry Datasets
6.3.3 Interpolation Between Cytometry Datasets
6.3.3.1 Gate Transportation
6.3.3.2 Reduction of Batch Effects
6.4 Conclusions
References
7 Derivation of Optimal Experimental Design Methods for Applications in Cytogenetic Biodosimetry
7.1 Introduction
7.2 Dose–Response Model
7.3 Optimal Experimental Design
7.4 OED for Cytogenetic Biodosimetry
7.5 Applied Examples
7.5.1 Dicentric Plus Ring Chromosomes
7.5.2 Translocation Assay
7.6 Conclusions
Code
Author Contribution Statement
References
8 Multiple Imputation for Compositional Data (MICoDa) Adjusting for Covariates
8.1 Introduction
8.2 MICoDa Methodology
8.3 Performance of MICoDa Using a Simulation Study
8.4 Illustration
8.4.1 Children\'s Activity Data
8.4.2 Diversity in Microbial Compositions
8.5 Discussion
Appendix 1: Table of Acronyms with Summary Details
Appendix 2: Proof of Lemma 1
Appendix 3: CLME Trend Analysis Based on Multiple Linear Regression (Traditional Model)
References
9 Integral Analysis of Circadian Rhythms
9.1 Introduction
9.2 Techniques Used to Measure Circadian Rhythms
9.2.1 Motor Activity
9.2.1.1 In Animal Models: General Motor Activity and Wheel Running Activity
9.2.1.2 In Humans: Actigraphy
9.2.2 Thermometry
9.2.3 Light Exposure
9.2.4 Multivariable Ambulatory Circadian Monitoring
9.2.5 Feeding Rhythms (Diary, Event Marker, Animals)
9.2.6 Hormone Determination (Sampling, Analytic Techniques, Melatonin, DLMO)
9.2.7 Gene Expression
9.3 Importance of the Frequency
9.4 Determinism in Time Series
9.5 Analysis of Rhythmicity
9.5.1 First Step: Filtering the Data
9.5.2 Parametric Methods to Determine the Existence of Circadian Rhythmicity
9.5.2.1 Graphic Analysis
9.5.2.2 Waveform Analysis
9.5.2.3 Periodogram
9.5.2.4 Based on Analysis of Variance
9.5.2.5 Cosinor Method
9.5.2.6 Fourier Analysis
9.5.3 Non-parametric Analysis
9.5.3.1 Parameters Related to Rhythm Amplitude
9.5.3.2 Parameters Related to the Rhythm Timing or Phase Markers
9.5.3.3 Parameters Indicating Regularity and Fragmentation of the Circadian Pattern
9.5.3.4 Parameters for Synchronization
9.5.3.5 Integrated Indexes That Provide Information About the Health and Robustness of the Circadian System
9.5.4 Intrinsic Circular Nature of the Data
9.6 Dynamic Circadian Models
9.6.1 Inputs and Outputs: Phase Response Curves
9.6.2 Kronauer\'s Dynamic Model
9.6.2.1 Model Behavior and Parameters
9.6.2.2 Model Outputs
9.6.2.3 St. Hilaire\'s Version of the Model
9.6.3 Two-Process Model of Sleep Regulation
9.6.3.1 Model Behavior and Parameters
9.6.3.2 Parameter Values for Healthy Adults
9.6.4 Combining Circadian Oscillators and Sleep Models
9.6.5 Further Reading
9.7 Conclusion and Future Work
References
10 Modelling the Circadian Variation of Electrocardiographic Parameters with Frequency Modulated Models
10.1 Introduction
10.2 Methods
10.2.1 FMM Background
10.2.2 HRV Wave (HRV-WA) Algorithm
10.3 Results
10.3.1 HRV-WA Performance Modeling 24-h HRV. Cosinor Comparison
10.3.2 HRV-WA as a Tool for an Interpretable Characterization of HRV
10.4 Discussion
Appendix
References
11 Novel Modeling Proposals for the Analysis of Pattern Electroretinogram Signals
11.1 Introduction
11.2 Methodological Background
11.2.1 Basic Concepts
11.2.1.1 AM-FM Signal Decomposition
11.2.2 The FMM Model
11.2.3 Measure of Goodness of Fit
11.2.4 Identification and Estimation Algorithm for Incomplete Oscillatory Signals
11.2.5 FMM Model with Autoregressive Errors: The FMMmq1 Model
11.2.5.1 Likelihood Function
11.2.5.2 Maximum Likelihood Estimation
11.3 Results
11.3.1 Real Data Analysis
11.3.2 Simulation Experiments
11.3.2.1 Validation of the Estimation Approach for Incomplete Signals
11.3.2.2 Validation of FMMmq1 Models
11.4 Discussion
References
Index