University of Illinois Urbana-Champaign

Predicting learning success from patterns of pre-training magnetic resonance images

Vo, Loan

Content Files
Vo_Loan.pdf

Permalink

https://hdl.handle.net/2142/31976

Description

Title
Predicting learning success from patterns of pre-training magnetic resonance images
Author(s)
Vo, Loan
Issue Date
2012-06-27T21:22:40Z
Director of Research (if dissertation) or Advisor (if thesis)
Wang, Michelle Y.
Doctoral Committee Chair(s)
Liang, Zhi-Pei
Committee Member(s)
  • Wang, Michelle Y.
  • Kramer, Arthur F.
  • Ahuja, Narendra
  • Coleman, Todd P.
Department of Study
Electrical and Computer Engineering
Discipline
Electrical and Computer Engineering
Degree Granting Institution
University of Illinois at Urbana-Champaign
Degree Name
Ph.D.
Degree Level
Dissertation
Date of Ingest
2012-06-27T21:22:40Z
Keyword(s)
  • nonheme iron
  • T2*
  • time-averaged T2*
  • Magnetic resonance imaging (MRI)
  • learning
  • striatum
  • caudate nucleus
  • putamen
  • nucleus accumbens
Abstract
Performance in most complex cognitive and psychomotor tasks improves with training, yet the extent of improvement varies among individuals. Is it possible to forecast the benefit that a person might reap from training? What is the mechanism underlying learning? Several behavioral measures have been used to predict individual differences in task improvement, but their predictive power is limited. Our multi-voxel pattern analysis (support vector regression) of the time-averaged blood oxygen level dependent (BOLD) brain activity in the dorsal but not the ventral striatum, recorded before training, predicts subsequent learning success with high accuracy. The fact that the high prediction accuracy of the data did not depend on the task subjects were performing during the recording might suggest that individual differences in neuroanatomy or persistent physiology predict whether and to what extent people will benefit from training in a complex task. To find out the physiology behind the possibility of predicting learning from time-averaged T2*-weighted images, a follow-up experiment was designed and performed with additional magnetic resonance (MR) measurements, including susceptibility-sensitive ones, such as susceptibility-weighted imaging (SWI), T2-, T2*-quantitative as well as diffusion tensor imaging (DTI) and arterial spin labeling (ASL). We then discovered that (patterns of) nonheme iron (not heme) is the underlying factor driving learning prediction. This discovery of the relationship between iron concentrations and learning ability in healthy young adults could not only guide the development of potential neuromarkers for a person's memory and executive control functions, but also help design customized learning-interventions to improve cognition or prevent its decline.
Graduation Semester
2012-05
Permalink
http://hdl.handle.net/2142/31976
Copyright and License Information
Copyright 2012 Loan Vo

Owning Collections

Graduate Dissertations and Theses at Illinois PRIMARY
Graduate Theses and Dissertations at Illinois
Dissertations and Theses - Electrical and Computer Engineering
Dissertations and Theses in Electrical and Computer Engineering