Neural Engineering Techniques for Autism Spectrum Disorder, Volume 2: Diagnosis and Clinical Analysis

Front Cover
Neural Engineering Techniques for Autism Spectrum Disorder, Volume 2
Copyright Page
Dedication
Contents
List of contributors
About the editors
Acknowledgments
1 Autism and clinical analysis: Diagnosis
	1 Remote telehealth assessments for autism spectrum disorder
		1.1 Introduction
			1.1.1 In-person standardized assessments for autism spectrum disorder
			1.1.2 Significance of remote assessments for autism spectrum disorder
		1.2 Telehealth assessments
			1.2.1 Videoconferencing (live/in vivo)
			1.2.2 Asynchronous video analysis: current
			1.2.3 Asynchronous video analysis: retrospective
			1.2.4 Mobile applications
			1.2.5 Online websites
			1.2.6 Other forms of technology
		1.3 Implications
			1.3.1 Future directions
		References
	2 Maternal immune dysregulation and autism spectrum disorder
		2.1 Introduction
		2.2 Cytokines and chemokines (overview)
			2.2.1 Cytokines and chemokines in the central nervous system
			2.2.2 Effect of cytokine/chemokine production in brain development
				2.2.2.1 Cytokines and chemokines in brain function
				2.2.2.2 Immune mediators and brain development
			2.2.3 Maternal immune dysregulation and developmental outcomes of offspring
			2.2.4 Maternal immune activation and autism spectrum disorder
			2.2.5 Maternal stress and autism spectrum disorder
			2.2.6 Maternal gut microbiome and autism spectrum disorder
			2.2.7 Alterations in cytokine and chemokine profiles during gestation and the neonatal period
				2.2.7.1 Other cytokines
		2.3 Autoantibodies reactive to brain antigens
			2.3.1 Autoantibody overview
			2.3.2 Autoantibodies and brain pathologies
			2.3.3 Autoantibodies and autism spectrum disorder
			2.3.4 Maternal autoantibodies and neurodevelopmental alterations
			2.3.5 Maternal autoantibody-related autism spectrum disorder overview
			2.3.6 MAR ASD animal models
			2.3.7 Maternal autoantibody-related fetal- brain targets and autism spectrum disorder
			2.3.8 Maternal autoantibodies as potential autism spectrum disorder-risk biomarkers
		2.4 Concluding remarks
		References
	3 Reading differences in eye-tracking data as a marker of high-functioning autism in adults and comparison to results from ...
		3.1 Introduction
		3.2 Related work
		3.3 Automated detection of high-functioning autism in adults with eye-tracking data from web tasks
		3.4 The proposed approach
			3.4.1 Data collection
			3.4.2 Participants
			3.4.3 Stimuli and tasks
			3.4.4 Apparatus
			3.4.5 Procedure
			3.4.6 Data preprocessing
		3.5 Experiments
		3.6 Results
		3.7 Discussion
		3.8 Conclusion
		3.9 Open data
		References
	4 Parents of children with autism spectrum disorders: interventions with and for them
		4.1 Introduction
		4.2 Parent participation in early comprehensive intervention programs
			4.2.1 Parental training
			4.2.2 Pivotal Response Training Program
			4.2.3 Treatment and Education of Autistic related Communication Handicapped Children Program
			4.2.4 Early Start Denver Model
		4.3 Programs for the development of parent–child interaction
			4.3.1 Hanen’s more than words
			4.3.2 Preschool autism communication trial
			4.3.3 Joint Attention Symbolic Play, Engagement, and Regulation
			4.3.4 Improving Parents as Communication Teachers
			4.3.5 Parent–child interaction therapy
			4.3.6 Stepping Stones Triple P
		4.4 Parent–child intervention based on anxiety reduction
			4.4.1 Cognitive behavioral therapy for anxiety reduction in children with autism spectrum disorders with parental intervention
			4.4.2 Mindfulness-based intervention
				4.4.2.1 Mindfulness training for parents mindfulness parenting
				4.4.2.2 Mindfulness training for Youngsters with autism spectrum disorder
		4.5 Conclusion
		References
	5 Applications of machine learning methods to assist the diagnosis of autism spectrum disorder
		5.1 Introduction
		5.2 Background and related work
			5.2.1 Analysis of visual attention in autism
			5.2.2 Machine learning for autism diagnosis
		5.3 Data description
			5.3.1 Participants
			5.3.2 Experimental protocol
			5.3.3 Visualization of eye-tracking scanpaths
		5.4 Unsupervised learning: clustering of eye-tracking scanpaths
			5.4.1 Image preprocessing
			5.4.2 Feature extraction using principal component analysis and t-SNE
			5.4.3 Feature extraction using deep autoencoder
			5.4.4 K-Means clustering
			5.4.5 Quality of clusters
			5.4.6 Cluster analysis
		5.5 Supervised learning: classification model
			5.5.1 Data preprocessing and augmentation
			5.5.2 Model design
			5.5.3 Classification accuracy
		5.6 Demo application
		5.7 Limitations
		5.8 Conclusions
		References
	6 Potential approaches and recent advances in biomarker discovery in autism spectrum disorders
		6.1 Introduction
		6.2 Diagnosis and categories of biomarkers
			6.2.1 Human brain connectome: structural, functional, and molecular neuroimaging biomarkers for autism spectrum disorder
			6.2.2 Molecular biomarkers
				6.2.2.1 Genes involved in autism spectrum disorder
				6.2.2.2 Epigenetic regulation
				6.2.2.3 Transcriptomic profiling
				6.2.2.4 Noncoding RNAs profiling
				6.2.2.5 Microparticles and extracellular vesicles
				6.2.2.6 Proteomics
					6.2.2.6.1 Proteins involved in neurodevelopment and function impairment
					6.2.2.6.2 Proteins involved in lipid metabolism
					6.2.2.6.3 Proteins involved in immunology, complement system, and inflammation
					6.2.2.6.4 Phosphoproteomics
				6.2.2.7 Metabolomics
					6.2.2.7.1 Neurotransmission-related metabolites
					6.2.2.7.2 Abnormal amino acid and fatty acid-related metabolites
					6.2.2.7.3 Gut microbiota-related metabolites
				6.2.2.8 Mitochondria dysfunction
			6.2.3 Maternal and paternal biomarkers: pregnancy and its potential association with ASD
				6.2.3.1 Significance of prenatal components in autism spectrum disorder risk development
				6.2.3.2 Screening for prenatal hormones as predictive biomarkers of autism spectrum disorder
				6.2.3.3 Screening for prenatal inflammatory cytokines as predictive biomarkers of autism spectrum disorder
				6.2.3.4 Autoantibodies
				6.2.3.5 Screening for metabolites as predictive biomarkers of autism spectrum disorder
				6.2.3.6 Sperm DNA methylation
				6.2.3.7 Biomarker discovery, drug development, and personalized medicine
			6.2.4 Next generation of diagnostic biomarkers
				6.2.4.1 Artificial intelligence advancing diagnosis
				6.2.4.2 Artificial intelligence and autism spectrum disorder diagnosis: eye-tracking
				6.2.4.3 Artificial intelligence-based systems biology approaches in multiomics
		6.3 Conclusion
		References
	7 Detection and identification of warning signs of autism spectrum disorder: instruments and strategies for its application
		7.1 Introduction
		7.2 Importance of early detection
		7.3 Differential diagnosis
			7.3.1 A brief history of the relationship between autism and psychosis
			7.3.2 Similarities
			7.3.3 Distinguishing features
		7.4 Detection and screening process
		7.5 Symptom detection vs Diagnosis
		7.6 Impact on the family of detecting and diagnosing Autism Spectrum Disorder
		7.7 Choice of screening instruments according to age of application and cultural environment of implementation
		7.8 Discussion
		7.9 Conclusions
		References
	8 Machine learning in autism spectrum disorder diagnosis and treatment: techniques and applications
		8.1 Introduction
		8.2 Utilizing machine learning algorithms to diagnose autism spectrum disorder
			8.2.1 Dataset with behavioral characteristics
			8.2.2 Dataset with personal/cognitive characteristics
			8.2.3 Recommendations
		8.3 Feature analysis
			8.3.1 Dimensionality reduction
			8.3.2 Feature representation
			8.3.3 Recommendations
		8.4 Technological applications
		8.5 Conclusion
		References
	9 Inhibition of lysine-specific demethylase 1 enzyme activity by TAK-418 as a novel therapy for autism
		9.1 Introduction
		9.2 Lysine-specific demethylase 1 as the potential therapeutic target for autism spectrum disorder
			9.2.1 Druggability in targeting epigenetic factors
			9.2.2 Potential therapeutic functions of lysine-specific demethylase 1 inhibition
			9.2.3 Concern regarding the on-target toxicity of general lysine-specific demethylase 1 inhibitors
		9.3 Discovery of the “enzyme activity-specific” inhibitors of lysine-specific demethylase 1
			9.3.1 Original screening flow
			9.3.2 Discovery of T-448 and TAK-418
			9.3.3 Unique inhibitory mechanism of T-448 and TAK-418
			9.3.4 Low risk of hematological toxicity by T-448 and TAK-418 in rodents
			9.3.5 Preclinical efficacy of T-448 and TAK-418
			9.3.6 Hypothesis of the mechanism of action of T-448 and TAK-418
		9.4 Discussion
		9.5 Conclusion
		References
	10 Behavioral phenotype features of autism
		10.1 Introduction
		10.2 Eye movement behavior phenotype of autism
			10.2.1 Natural stimuli
				10.2.1.1 Dataset
				10.2.1.2 Analysis
				10.2.1.3 Gaze pattern classification and saliency prediction
				10.2.1.4 Models submitted to Saliency4ASD
				10.2.1.5 Evaluation criteria
				10.2.1.6 Results of Saliency4ASD
			10.2.2 Face stimuli
				10.2.2.1 Dataset
				10.2.2.2 Analysis
				10.2.2.3 Methods and results
			10.2.3 Gaze-following stimuli
				10.2.3.1 Dataset
				10.2.3.2 Analysis
				10.2.3.3 Methods and results
		10.3 Action behavior phenotype
			10.3.1 Dataset and analysis
			10.3.2 Methods and results
		10.4 Drawing behavior phenotype
			10.4.1 Dataset
			10.4.2 Analysis
			10.4.3 Results and discussion
		10.5 Discussion and conclusion
		References
	11 Development of an animated infographic about autistic spectrum disorder
		11.1 Introduction
		11.2 Infographics
			11.2.1 Study population
			11.2.2 Development
			11.2.3 Validation and testing
		11.3 Results
		11.4 Discussion
		11.5 Conclusion
		References
	12 Fundamentals of machine-learning modeling for behavioral screening and diagnosis of autism spectrum disorder
		12.1 Introduction
		12.2 Current autism spectrum disorder screening and diagnostic practices
			12.2.1 Commonly used autism spectrum disorder screening instruments
			12.2.2 Common problems with current autism spectrum disorder screening and diagnostic practices
		12.3 Machine learning-based assessment of autism spectrum disorder
			12.3.1 Commonly used datasets for machine learning-based behavioral assessment of autism spectrum disorder
			12.3.2 Dimensionality reduction
			12.3.3 Commonly used dimensionality reduction techniques
				12.3.3.1 Trial-and-error technique
				12.3.3.2 Feature selection techniques
				12.3.3.3 Feature transformation techniques
			12.3.4 Classification algorithms
			12.3.5 Model selection
				12.3.5.1 Decision trees
				12.3.5.2 K-nearest neighbor
				12.3.5.3 Naïve Bayes
				12.3.5.4 Logistic regression
				12.3.5.5 Support vector machine
			12.3.6 Confusion matrix
				12.3.6.1 Sensitivity and specificity
		12.4 Conclusion
		References
	13 A comprehensive study on atlas-based classification of autism spectrum disorder using functional connectivity features f...
		13.1 Introduction
		13.2 Overview of functional magnetic resonance imaging
			13.2.1 Clinical application
		13.3 Literature review
			13.3.1 Structural magnetic resonance imaging-based autism detection
			13.3.2 Functional magnetic resonance imaging-based autism detection
			13.3.3 Structural and functional magnetic resonance imaging-based autism detection
		13.4 Materials and methods
			13.4.1 Preprocessing
			13.4.2 Blood oxygen level dependent time-series signal extraction from four dimensional functional magnetic resonance imagi...
				13.4.2.1 Automated anatomical labeling
				13.4.2.2 Bootstrap analysis of stable clusters
				13.4.2.3 Craddock 200
				13.4.2.4 Craddock 400
				13.4.2.5 Power
				13.4.2.6 Eickhoff–Zilles
				13.4.2.7 Harvard–Oxford
				13.4.2.8 Talaraich and tournoux
				13.4.2.9 Dosenbach 160
				13.4.2.10 Massive online dictionary learning
			13.4.3 Building functional connectivity matrix
				13.4.3.1 Full correlation/Pearson’s correlation
				13.4.3.2 Partial correlation
				13.4.3.3 Tangent space embedding
			13.4.4 Feature vector
			13.4.5 Classification
				13.4.5.1 Hidden layers and neurons per layer
				13.4.5.2 Activation function
				13.4.5.3 Optimizer
				13.4.5.4 Loss function
				13.4.5.5 Regularizer
					13.4.5.5.1 L2 regularizer
					13.4.5.5.2 Dropout
				13.4.5.6 Batch size
				13.4.5.7 Callback functions
					13.4.5.7.1 Early stopping
					13.4.5.7.2 Model checkpoint
		13.5 Experimental results and analysis
			13.5.1 Dataset description
			13.5.2 Evaluation of autism spectrum disorder detection framework
			13.5.3 Performance evaluation using model-2
		13.6 Conclusion
		13.7 Future work
		References
	14 Event-related potentials and gamma oscillations in EEG as functional diagnostic biomarkers and outcomes in autism spectr...
		14.1 Introduction
		14.2 Neurophysiological biomarkers
			14.2.1 Introduction to event-related potentials and evoked brain waves oscillations
			14.2.2 Rationale for approach using EEG/ERP measures in studying attention in ASD
			14.2.3 Visual oddball task with illusory figures
			14.2.4 ERP data acquisition and signal processing
			14.2.5 Event-related potentials in autism and ADHD
			14.2.6 ERP measures in illusory figure (Kanizsa) categorization task
			14.2.7 Motor preparation deficits in ASD
			14.2.8 ERP in Posner cued spatial attention task
			14.2.9 Lateralized Readiness Potential (LRP) as an index of motor preparation in ASD and ADHD
		14.3 Gamma oscillations as potential neuromarkers in neurodevelopmental disorders
			14.3.1 Gamma oscillations
			14.3.2 Cortical excitation/inhibition (E/I) bias and brainwave oscillations
			14.3.3 Gamma oscillations in ASD
			14.3.4 Hemispheric asymmetry of gamma
		14.4 ERP and induced gamma oscillations in facial categorization task in ASD, ADHD, and TD groups
			14.4.1 ERP results in ToM task
			14.4.2 Induced gamma analysis and results in ToM task
		14.5 Evoked and induced EEG data acquisition and processing in Kanizsa oddball task
		14.6 Conclusions
		References
Index
Back Cover