Search

Scholarly Works (29 results)

Sort By:

Show:

Thesis
Peer Reviewed

Modeling Time-varying Trends in ERP Data with Applications to an Implicit Learning Paradigm in Autism

Hasenstab, Kyle Andrew
Advisor(s): Senturk, Damla

UCLA Electronic Theses and Dissertations (2015)

Event-related potential (ERP) studies are a subset of experimental frameworks within the field of electroencephalography (EEG) that focus on ERPs, the electrical potential outputted by a subject's brain when presented with an implicit task in the form of stimuli. Data comprise an ERP repetition observed for each stimulus across electrodes on the scalp, producing a complex data structure consisting of a functional, longitudinal and spatial dimension. In typical ERP studies, the dimension of data is reduced into a single measure for each subject by cross-sectionally averaging ERP across longitudinal and spatial repetitions. Features are then extracted from the averaged ERP and analyzed using simple statistical methods, ignoring additional information that may be found in the collapsed dimensions. In this dissertation, three types of methodology are proposed for preserving and analyzing the lost dimensions of ERP data. The first method, moving average processed ERP (MAP-ERP), is a two-step approach comprised of a meta-preprocessing step to preserve longitudinal information and a weighted mixed effects regression framework to allow modeling of the resulting meta-preprocessed data. The proposed robust functional clustering (RFC) algorithm identifies substructures in features of the longitudinal ERP processes while accounting for subject-level covariance heterogeneity induced by meta-preprocessing. Finally, the proposed multidimensional functional principal components analysis (MD-FPCA) utilizes a two-stage procedure to summarize important characteristics across all three dimensions of the ERP data structure into an interpretable, low-dimensional form. Proposed methods are applied to a study on neural correlates of visual implicit learning in young children with autism spectrum disorder (ASD). Applications of the proposed methods reveal meaningful trends and substructures in the implicit learning processes of ASD children when compared to typically developing controls. Results indicate proposed methodology effectively preserves important information contained within the multiple dimensions of ERP.

Cover page: Modeling Time-varying Trends in ERP Data with Applications to an Implicit Learning Paradigm in Autism

Thesis
Peer Reviewed

Statistical Directions for the Analysis of Participatory Mobile Health Asthma Management Data

Hasenstab, Kyle Andrew
Advisor(s): Hansen, Mark H

UCLA Electronic Theses and Dissertations (2012)

Mobile technology has become increasingly popular in the past decade through the combination of device portability and advances in network and internet technology. Smart phones, in particular, are at the forefront of these technologies, enabling users to remotely track and further involve themselves in the management of personal health through participatory mobile health applications. Thus far, the majority of participatory applications currently provide users with data visualizations displaying information germane to the user's medical condition, however, there is still a need for in-depth exploratory and inference-based data analysis using advanced statistical methods to maximize the discernment of potential implications carried in these data streams. This paper provides an overview of the structure of participatory data using an asthma management application as an initial platform and discusses several directions for statistical analysis motivated by three usage cases: individuals using the application, creators of the application, and the scientific community. Methods include functional and semi-parametric data analysis, mixed modeling, and clustering methods to model variables representing asthma wellness as a function of subject-specific, population level, and latent spatiotemporal factors. Societal implications are also discussed.

Cover page: Statistical Directions for the Analysis of Participatory Mobile Health Asthma Management Data

Article
Peer Reviewed

CNN-based Deformable Registration Facilitates Fast and Accurate Air Trapping Measurements at Inspiratory and Expiratory CT

UC San Diego Previously Published Works (2022)

Purpose

To develop a convolutional neural network (CNN)-based deformable lung registration algorithm to reduce computation time and assess its potential for lobar air trapping quantification.

Materials and methods

In this retrospective study, a CNN algorithm was developed to perform deformable registration of lung CT (LungReg) using data on 9118 patients from the COPDGene Study (data collected between 2007 and 2012). Loss function constraints included cross-correlation, displacement field regularization, lobar segmentation overlap, and the Jacobian determinant. LungReg was compared with a standard diffeomorphic registration (SyN) for lobar Dice overlap, percentage voxels with nonpositive Jacobian determinants, and inference runtime using paired t tests. Landmark colocalization error (LCE) across 10 patients was compared using a random effects model. Agreement between LungReg and SyN air trapping measurements was assessed using intraclass correlation coefficient. The ability of LungReg versus SyN emphysema and air trapping measurements to predict Global Initiative for Chronic Obstructive Lung Disease (GOLD) stages was compared using area under the receiver operating characteristic curves.

Results

Average performance of LungReg versus SyN showed lobar Dice overlap score of 0.91-0.97 versus 0.89-0.95, respectively (P < .001); percentage voxels with nonpositive Jacobian determinant of 0.04 versus 0.10, respectively (P < .001); inference run time of 0.99 second (graphics processing unit) and 2.27 seconds (central processing unit) versus 418.46 seconds (central processing unit) (P < .001); and LCE of 7.21 mm versus 6.93 mm (P < .001). LungReg and SyN whole-lung and lobar air trapping measurements achieved excellent agreement (intraclass correlation coefficients > 0.98). LungReg versus SyN area under the receiver operating characteristic curves for predicting GOLD stage were not statistically different (range, 0.88-0.95 vs 0.88-0.95, respectively; P = .31-.95).

Conclusion

CNN-based deformable lung registration is accurate and fully automated, with runtime feasible for clinical lobar air trapping quantification, and has potential to improve diagnosis of small airway diseases.Keywords: Air Trapping, Convolutional Neural Network, Deformable Registration, Small Airway Disease, CT, Lung, Semisupervised Learning, Unsupervised Learning Supplemental material is available for this article. © RSNA, 2021 An earlier incorrect version of this article appeared online. This article was corrected on December 22, 2021.

Cover page: CNN-based Deformable Registration Facilitates Fast and Accurate Air Trapping Measurements at Inspiratory and Expiratory CT

Article
Peer Reviewed

Robust functional clustering of ERP data with application to a study of implicit learning in autism

UC Irvine Previously Published Works (2016)

Motivated by a study on visual implicit learning in young children with Autism Spectrum Disorder (ASD), we propose a robust functional clustering (RFC) algorithm to identify subgroups within electroencephalography (EEG) data. The proposed RFC is an iterative algorithm based on functional principal component analysis, where cluster membership is updated via predictions of the functional trajectories obtained through a non-parametric random effects model. We consider functional data resulting from event-related potential (ERP) waveforms representing EEG time-locked to stimuli over the course of an implicit learning experiment, after applying a previously proposed meta-preprocessing step. This meta-preprocessing is designed to increase the low signal-to-noise ratio in the raw data and to mitigate the longitudinal changes in the ERP waveforms which characterize the nature and speed of learning. The resulting functional ERP components (peak amplitudes and latencies) inherently exhibit covariance heterogeneity due to low data quality over some stimuli inducing the averaging of different numbers of waveforms in sliding windows of the meta-preprocessing step. The proposed RFC algorithm incorporates this known covariance heterogeneity into the clustering algorithm, improving cluster quality, as illustrated in the data application and extensive simulation studies. ASD is a heterogeneous syndrome and identifying subgroups within ASD children is of interest for understanding the diverse nature of this complex disorder. Applications to the implicit learning paradigm identify subgroups within ASD and typically developing children with diverse learning patterns over the course of the experiment, which may inform clinical stratification of ASD.

Cover page: Robust functional clustering of ERP data with application to a study of implicit learning in autism

Article
Peer Reviewed

Simulating clinical features on chest radiographs for medical image exploration and CNN explainability using a style-based generative adversarial autoencoder.

UC San Diego Previously Published Works (2024)

Explainability of convolutional neural networks (CNNs) is integral for their adoption into radiological practice. Commonly used attribution methods localize image areas important for CNN prediction but do not characterize relevant imaging features underlying these areas, acting as a barrier to the adoption of CNNs for clinical use. We therefore propose Semantic Exploration and Explainability using a Style-based Generative Adversarial Autoencoder Network (SEE-GAAN), an explainability framework that uses latent space manipulation to generate a sequence of synthetic images that semantically visualizes how clinical and CNN features manifest within medical images. Visual analysis of changes in these sequences then facilitates the interpretation of features, thereby improving explainability. SEE-GAAN was first developed on a cohort of 26,664 chest radiographs across 15,409 patients from our institution. SEE-GAAN sequences were then generated across several clinical features and CNN predictions of NT-pro B-type natriuretic peptide (BNPP) as a proxy for acute heart failure. Radiological interpretations indicated SEE-GAAN sequences captured relevant changes in anatomical and pathological morphology associated with clinical and CNN predictions and clarified ambiguous areas highlighted by commonly used attribution methods. Our study demonstrates SEE-GAAN can facilitate our understanding of clinical features for imaging biomarker exploration and improve CNN transparency over commonly used explainability methods.

Cover page: Simulating clinical features on chest radiographs for medical image exploration and CNN explainability using a style-based generative adversarial autoencoder.

Article
Peer Reviewed

Physiologic artifacts in resting state oscillations in young children: methodological considerations for noisy data

UCLA Previously Published Works (2015)

We quantified the potential effects of physiologic artifact on the estimation of EEG band power in a cohort of typically developing children in order to guide artifact rejection methods in quantitative EEG data analysis in developmental populations. High density EEG was recorded for 2 min while children, ages 2-6, watched a video of bubbles. Segments of data were categorized as blinks, saccades, EMG or artifact-free, and both absolute and relative power in the theta (4-7 Hz), alpha (8-12 Hz), beta (13-30 Hz) and gamma (35-45 Hz) bands were calculated in 9 regions for each category. Using a linear mixed model approach with artifact type, region and their interaction as predictors, we compared mean band power between clean data and each type of artifact. We found significant differences in mean relative and absolute power between artifacts and artifact-free segments in all frequency bands. The magnitude and direction of the differences varied based on power type, region, and frequency band. The most significant differences in mean band power were found in the gamma band for EMG artifact and the theta band for ocular artifacts. Artifact detection strategies need to be sensitive to the oscillations of interest for a given analysis, with the most conservative approach being the removal of all EMG and ocular artifact from EEG data. Quantitative EEG holds considerable promise as a clinical biomarker of both typical and atypical development. However, there needs to be transparency in the choice of power type, regions of interest, and frequency band, as each of these variables are differentially vulnerable to noise, and therefore, their interpretation depends on the methods used to identify and remove artifacts.

Cover page: Physiologic artifacts in resting state oscillations in young children: methodological considerations for noisy data

Article
Peer Reviewed

Identifying Longitudinal Trends within EEGExperiments

Research Reports (2015)

Article
Peer Reviewed

A Multi-Dimensional Functional Principal Components Analysis of EEG Data

UCLA Previously Published Works (2017)

The electroencephalography (EEG) data created in event-related potential (ERP) experiments have a complex high-dimensional structure. Each stimulus presentation, or trial, generates an ERP waveform which is an instance of functional data. The experiments are made up of sequences of multiple trials, resulting in longitudinal functional data and moreover, responses are recorded at multiple electrodes on the scalp, adding an electrode dimension. Traditional EEG analyses involve multiple simplifications of this structure to increase the signal-to-noise ratio, effectively collapsing the functional and longitudinal components by identifying key features of the ERPs and averaging them across trials. Motivated by an implicit learning paradigm used in autism research in which the functional, longitudinal, and electrode components all have critical interpretations, we propose a multidimensional functional principal components analysis (MD-FPCA) technique which does not collapse any of the dimensions of the ERP data. The proposed decomposition is based on separation of the total variation into subject and subunit level variation which are further decomposed in a two-stage functional principal components analysis. The proposed methodology is shown to be useful for modeling longitudinal trends in the ERP functions, leading to novel insights into the learning patterns of children with Autism Spectrum Disorder (ASD) and their typically developing peers as well as comparisons between the two groups. Finite sample properties of MD-FPCA are further studied via extensive simulations.

Article
Peer Reviewed