Marsman, J.B.C. (2006) Using a Support-Vector Machine for the analysis of fMRI images in a tactile frequency-discrimination task. Master's Thesis / Essay, Artificial Intelligence.
|
Text
AI_Ma_2006_JBCMarsman.CV.pdf - Published Version Download (1MB) | Preview |
Abstract
Traditionally, fMRI analysis uses a univariate method for statistical testing, such as the General Linear Model. In the past few years multivariate approaches have been developed also taking spatial relations between voxels into account, including Support-Vector Machines (SVMs). An SVM is a binary classifier which calculates the optimal hyperplane (decision boundary) between two classes in a training set. This hyperplane is then used to predict the class of the items in a test set. The goal of this study was to find correlates between human discrimination ability and the performance of the SVM on the captured IMRI images from this discrimination task. The task was to discriminate between different vibrotactile frequencies, where it is still an open question as to whether neural firing rate or synchronicity is used. fMRI signal intensity changes as a function of firing rate, but not synchronicity, and so could inform this question. We conducted an IMRI experiment where vibrotactile stimuli were applied to the right index finger. consisting of a reference frequency and a stimulus frequency. After each pair of stimuli, the subject had to respond with either higher or lower, using a button box held in the left hand. Beside standard fMRI preprocessing and stimulus-related averaging over each session, we used either Singular Value Decomposition or feature selection methods to decrease computational time. The results led to the conclusion that frequency is not spatially represented. This still leaves rate encoding and periodicity encoding to be possible representations of frequency.
| Item Type: | Thesis (Master's Thesis / Essay) |
|---|---|
| Degree programme: | Artificial Intelligence |
| Thesis type: | Master's Thesis / Essay |
| Language: | English |
| Date Deposited: | 15 Feb 2018 07:30 |
| Last Modified: | 15 Feb 2018 07:30 |
| URI: | https://fse.studenttheses.ub.rug.nl/id/eprint/9088 |
Actions (login required)
 |
View Item |