Measuring Glial Metabolism in Repetitive Brain Trauma and Alzheimer's Disease
Abstract
The most significant finding in this report was the application of denoising methods to multinuclear spectroscopy data which showed greatly improved signal-to-noise ratios of up to3.4 times that of the original data. Six methods: Single value decomposition (SVD), wavelet,sliding window, sliding window with Gaussian weighting, spline and spectral improvements with Fourier thresholding were tested to determine which provided optimal performance. The sliding window with Gaussian weighting showed the most consistent performance across alllevels of SNR with the best performance using spline and unweighted sliding window at low SNRdata. This analysis will be incorporated into our post-processing pipeline for the 13C spectroscopy data that will be acquired in this study after human subjects approval.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 2016
- Accession Number
- AD1034285
Entities
People
- Alexander Lin