Magnetic resonance imaging of neuronal activity is one of the most promising techniques in modern psychiatric research. While clear functional links with phenotypic variables have been established and detailed networks of activity robustly identified, fMRI scans have not yet yielded the robust biomarkers of psychiatric diseases, such as schizophrenia, which would allow for their use as a clinical diagnostic tool. One possible explanation for the lack of such results is that neural activity is highly non- stationary, whereas most analysis techniques assume that signal properties remain relatively static over time. Time-frequency analysis is a family of analytic techniques which do not assume that data is stationary, and thus is well suited to the analysis of neural time series. Resting state fMRI scans from a publicly available dataset were decomposed using the Wavelet transform and Hilbert Huang Transform, techniques from time-frequency analysis. The results of these processes were then used as the basis for calculating several properties of the fMRI signal within each voxel. The wavelet transform, a simpler technique, generated measures which showed broad differences between patients with schizophrenia and healthy controls but failed to reach statistical significance in the vast majority of situations. The Hilbert Huang transform, in contrast, showed significant increases in certain measures throughout areas associated with sensory processing, dysfunction in which is a symptom of schizophrenia. These results support the use of analysis techniques able to capture the nonstationarities in neural data and encourages the use of such techniques to explore the nature of the neural differences in psychiatric disorders.
| Identifer | oai:union.ndltd.org:bu.edu/oai:open.bu.edu:2144/44000 |
| Date | 09 March 2022 |
| Creators | Buck, Samuel Peter |
| Contributors | Thomas, Kevin C. |
| Source Sets | Boston University |
| Language | en_US |
| Detected Language | English |
| Type | Thesis/Dissertation |
| Rights | Attribution-NonCommercial 4.0 International, http://creativecommons.org/licenses/by-nc/4.0/ |
Page generated in 0.0017 seconds