Computational Learning Approaches to Data Analytics in Biomedical Applications

Cover
Computational Learning Approaches to Data Analytics in Biomedical Applications
Copyright
Preface and Acknowledgements
1 - Introduction
	References
2 - Data preprocessing
	2.1 Introduction
	2.2 Data preparation
		2.2.1 Initial cleansing
		2.2.2 Data imputation and missing values algorithms
			2.2.2.1 Removal Methods
			2.2.2.2 Utilization methods
			2.2.2.3 Maximum likelihood
		2.2.3 Imputation methods
			2.2.3.1 Single imputation methods
				2.2.3.1.1 Mean imputation
				2.2.3.1.2 Substitution of related observations
				2.2.3.1.3 Random selection
				2.2.3.1.4 Weighted K-nearest neighbors (KNN) imputation
			2.2.3.2 Multiple imputation
		2.2.4 Feature enumeration
			2.2.4.1 Special cases of categorical data representation using COBRIT traumatic brain injury data as an example
		2.2.5 Detecting and removing redundant features
			2.2.5.1 Pearson correlation
			2.2.5.2 Spearman correlation
		2.2.6 Recoding categorical features
		2.2.7 Outlier detection
		2.2.8 Normalization
		2.2.9 Domain experts
		2.2.10 Feature selection and extraction
	2.3 Example
	2.4 Summary
	References
3 - Clustering algorithms
	3.1 Introduction
	3.2 Proximity measures
	3.3 Clustering algorithms
		3.3.1 Hierarchical clustering
		3.3.2 Density-based clustering
		3.3.3 Subspace clustering
			3.3.3.1 Basic subspace clustering
				3.3.3.1.1 Grid-based subspace clustering
				3.3.3.1.2 Window-based subspace clustering
				3.3.3.1.3 Density-based subspace clustering
			3.3.3.2 Advanced subspace clustering
				3.3.3.2.1 3D subspace clustering
		3.3.4 Squared error-based clustering
		3.3.5 Fuzzy clustering
		3.3.6 Evolutionary computational technology-based clustering
		3.3.7 Neural network–based clustering
		3.3.8 Kernel learning-based clustering
		3.3.9 Large-scale data clustering
		3.3.10 High-dimensional data clustering
		3.3.11 Sequential data clustering
			3.3.11.1 Proximity-based sequence clustering
			3.3.11.2 Feature-based sequence clustering
			3.3.11.3 Model-based sequence clustering
	3.4 Adaptive resonance theory
		3.4.1 Fuzzy ART
		3.4.2 Fuzzy ARTMAP
		3.4.3 BARTMAP
	3.5 Summary
	References
4 - Selected approaches to supervised learning
	4.1 Backpropagation and related approaches
		4.1.1 Backpropagation
		4.1.2 Backpropagation through time
	4.2 Recurrent neural networks
	4.3 Long short-term memory
	4.4 Convolutional neural networks and deep learning
		4.4.1 Structure of convolutional neural network
		4.4.2 Deep belief networks
		4.4.3 Variational autoencoders
	4.5 Random forest, classification and Regression Tree, and related approaches
	4.6 Summary
	References
5 - Statistical analysis tools
	5.1 Introduction
	5.2 Tools for determining an appropriate analysis
	5.3 Statistical applications in cluster analysis
		5.3.1 Cluster evaluation tools: analyzing individual features
			5.3.1.1 Hypothesis testing and the 2-sample t-test
			5.3.1.2 Summary of hypothesis testing steps and application to clustering
			5.3.1.3 One-way ANOVA
			5.3.1.4 χ2 test for independence
		5.3.2 Cluster evaluation tools: multivariate analysis of features
	5.4 Software tools and examples
		5.4.1 Statistical software tools
			5.4.1.1 Example: clustering autism spectrum disorder phenotypes
			5.4.1.2 Correlation analysis
			5.4.1.3 Cluster evaluation of individual features
			5.4.1.4 Summary of results
	5.5 Summary
	References
6 - Genomic data analysis
	6.1 Introduction
	6.2 DNA methylation
		6.2.1 Introduction
		6.2.2 DNA methylation technology
		6.2.3 DNA methylation analysis
		6.2.4 Clustering applications for DNA methylation data
	6.3 SNP analysis
		6.3.1 Association studies
		6.3.2 Clustering with family-based association test (FBAT) analysis
			6.3.2.1 Quality control filtering
			6.3.2.2 Family-based association testing
			6.3.2.3 Multiple testing
			6.3.2.4 Adjustments for small sample size
			6.3.2.5 Implementation and analysis of results
	6.4 Biclustering for gene expression data analysis
		6.4.1 Introduction to biclustering
		6.4.2 Commonly used biclustering methods
		6.4.3 Evolutionary-based biclustering methods
		6.4.4 BARTMAP: a neural network-based biclustering algorithm
		6.4.5 External and internal validation metrics related to biclustering
	6.5 Summary
	References
7 - Evaluation of cluster validation metrics
	7.1 Introduction
	7.2 Related works
	7.3 Background
		7.3.1 Commonly used internal validation indices
		7.3.2 External validation indices
		7.3.3 Statistical methods
	7.4 Evaluation framework
	7.5 Experimental results and analysis
	7.6 Ensemble validation paradigm
	7.7 Summary
	References
8 - Data visualization
	8.1 Introduction
	8.2 Dimensionality reduction methods
		8.2.1 Linear projection algorithms
			8.2.1.1 Principal component analysis
			8.2.1.2 Independent component analysis
		8.2.2 Nonlinear projection algorithms
			8.2.2.1 Isomap
			8.2.2.2 T-Distributed Stochastic Neighbor Embedding (t-SNE)
			8.2.2.3 LargeVis
			8.2.2.4 Self-organizing maps
			8.2.2.5 Visualization of commonly used biomedical data sets from the UCI machine learning repository ()
	8.3 Topological data analysis
	8.4 Visualization for neural network architectures
	8.5 Summary
	References
9 - Data analysis and machine learning tools in MATLAB and Python
	9.1 Introduction
	9.2 Importing data
		9.2.1 Reading data in MATLAB
			9.2.1.1 Interactive import function
			9.2.1.2 Reading data as formatted tables
			9.2.1.3 Reading data as cellular arrays
			9.2.1.4 Reading data as numerical arrays and matrices
				9.2.1.4.1 xlsread function
				9.2.1.4.2 Functions for reading in data from text files as numerical arrays/matrices
				9.2.1.4.3 Reading images in MATLAB
		9.2.2 Reading data in Python
			9.2.2.1 Overview of external libraries and modules for Python
			9.2.2.2 Opening files in Python
				9.2.2.2.1 Reading text files in Python
				9.2.2.2.2 read_csv() function
				9.2.2.2.3 Other read functions
		9.2.3 Handling big data in MATLAB
			9.2.3.1 How to create data stores in MATLAB
				9.2.3.1.1 read function
				9.2.3.1.2 readall function
				9.2.3.1.3 hasdata function
				9.2.3.1.4 partition function
				9.2.3.1.5 numpartitions function
			9.2.3.2 Tall arrays
	9.3 Data preprocessing
		9.3.1 Missing values handling
			9.3.1.1 Handling missing values during reading
			9.3.1.2 Finding and replacing missing values
		9.3.2 Normalization
			9.3.2.1 z-score
		9.3.3 Outliers detection
	9.4 Tools and functions for implementing machine learning algorithms
		9.4.1 Clustering
			9.4.1.1 k-means
			9.4.1.2 Gaussian mixture model
			9.4.1.3 Hierarchical clustering
			9.4.1.4 Self-organizing map
		9.4.2 Prediction and classification
			9.4.2.1 Machine learning workflow
				9.4.2.1.1 Data Preparation
				9.4.2.1.2 Fitting and predicting tools
			9.4.2.2 Multiclass support vector machines
			9.4.2.3 Neural network classifier
			9.4.2.4 Performance evaluation and cross-validation tools
		9.4.3 Features reduction and features selection tools in MATLAB
			9.4.3.1 Built-in feature selection method
			9.4.3.2 Sequential features selection
		9.4.4 Features reduction and features selection tools in Python
			9.4.4.1 Removing features with low variance
			9.4.4.2 Recursive feature elimination
	9.5 Visualization
		9.5.1 Multidimensional scaling
			9.5.1.1 Pairwise distance calculation function pdist
			9.5.1.2 Perform multidimensional scaling
		9.5.2 Principal component analysis
		9.5.3 Visualization functions
	9.6 Clusters and classification evaluation functions
		9.6.1 Cluster evaluation
		9.6.2 Classification models evaluation
			9.6.2.1 Confusion matrix confusionmat
	9.7 Summary
	References
Index
	A
	B
	C
	D
	E
	F
	G
	H
	I
	J
	K
	L
	M
	N
	O
	P
	Q
	R
	S
	T
	U
	V
	W
	X
Back Cover