Link to the University of Pittsburgh Homepage
Link to the University Library System Homepage Link to the Contact Us Form

Advances in Statistical and Machine Learning Methods for Image Data, with Application to Alzheimer's Disease

Ding, Tianyu (2020) Advances in Statistical and Machine Learning Methods for Image Data, with Application to Alzheimer's Disease. Doctoral Dissertation, University of Pittsburgh. (Unpublished)

This is the latest version of this item.

[img]
Preview
PDF
Submitted Version

Download (3MB) | Preview

Abstract

The revolutionary development of neuroimage technology allows for the generation of large-scale neuroimage data in modern medical studies. For example, structural magnetic resonance imaging (sMRI) is widely used in segmenting neurodegenerative regions in the brain and positron-emission tomography (PET) is commonly used by clinicians and researchers to quantify the severity of Alzheimer's disease.

In the first part of this dissertation, we build “OASIS-AD”, which is a supervised learning model based on a well-validated automated segmentation tool “OASIS” in multiple sclerosis (MS). OASIS-AD considers the specific challenges raised by WMH in Alzheimer's Disease (AD) to reduce false discoveries. We show that OASIS-AD performs better than several existing automated white matter hyperintensity segmentation approaches.

In the second part of this dissertation, we develop an interpretable penalized multivariate high-dimensional method for image-on-scalar regression that can be used for association studies between high-dimensional PET images and patients' scalar measures. This method overcomes the lack of interpretability in regularized regression after reduced-rank decomposition through a novel encoder-decoder based penalty to regularize interpretable image characteristics. Empirical properties of the proposed approach are examined and compared to existing methods in simulation studies and in the analysis of PET images from subjects in a study of Alzheimer's Disease.

In the third part of this dissertation, we developed ACU-Net, an efficient convolutional network for medical image segmentation. The proposed deep learning network overcomes the small sample size problem of training a deep neural network when used for medical image segmentation. It also decreases computation cost by increasing the effective degrees of freedom through data augmentation and the novel use of convolutional layers blocks to compress the model. We show that ACU-Net can achieve competitive performance while dramatically decreases the computation cost compared with modern CNNs.

Public health significance: This dissertation proposes new statistical and machine learning methods for two aging-related problems: (1) automatically segmenting white matter hyperintensity (WMH), a biomarker of neurodegenerative pathology, and (2) estimating the association between neurodegeneration pathology and vascular measures, which are important to aging population living quality and can be studied by clinical neuroimage data.


Share

Citation/Export:
Social Networking:
Share |

Details

Item Type: University of Pittsburgh ETD
Status: Unpublished
Creators/Authors:
CreatorsEmailPitt UsernameORCID
Ding, Tianyutid16@pitt.edutid160000-0001-6660-8463
ETD Committee:
TitleMemberEmail AddressPitt UsernameORCID
Committee ChairKrafty, Robertrkrafty@pitt.edu
Committee MemberTudorascu, Danadlt30@pitt.edu
Committee MemberCohen, Anncohenad@upmc.edu
Committee MemberAnderson, Stewartsja@pitt.edu
Date: 30 July 2020
Date Type: Publication
Defense Date: 20 May 2020
Approval Date: 30 July 2020
Submission Date: 25 May 2020
Access Restriction: No restriction; Release the ETD for access worldwide immediately.
Number of Pages: 82
Institution: University of Pittsburgh
Schools and Programs: Graduate School of Public Health > Biostatistics
Degree: PhD - Doctor of Philosophy
Thesis Type: Doctoral Dissertation
Refereed: Yes
Uncontrolled Keywords: Alzheimer's Disease; White Matter Hyperintensity; Logistic Regression; Spacial Information; Segmentation; ADMM; Image-on-Scalar Regression; Reduced-Rank Regression; Sparsity; Interpretability; Deep neural network; CNN
Date Deposited: 31 Jul 2020 03:06
Last Modified: 31 Jul 2020 03:06
URI: http://d-scholarship.pitt.edu/id/eprint/39222

Available Versions of this Item

  • Advances in Statistical and Machine Learning Methods for Image Data, with Application to Alzheimer's Disease. (deposited 31 Jul 2020 03:06) [Currently Displayed]

Metrics

Monthly Views for the past 3 years

Plum Analytics


Actions (login required)

View Item View Item