The use of bootstrapping to measure image differences in fMRI data
Averick, John Wesley
MetadataShow full item record
Schizophrenia is a severe mental disorder that affects millions of people and is subject to much research. Functional magnetic resonance imaging (fMRI) is a tool that provides, by observing changes in blood oxygenation, an indirect measure of brain activation. As schizophrenia is known to be partially hereditary, a study was conducted on patients, their relatives, and a control group in order to measure differences in brain activation patterns during the performance of an antisaccade task. In this master's thesis, the processes of bootstrapping and distribution construction are used in an attempt to assess the differences in brain activation among the three groups. The results indicate differences both between the control group and the relatives and between the controls and patients, but no definite evidence that the two differences coincide. Simulated data sets are used to confirm the efficacy of the methodology used. Speculation is given as to incongruities with previous analyses of the data.