Robust analysis of short echo time ¹H MRSI of human brain

Abstract

Short echo time proton MR Spectroscopic Imaging (MRSI) suffers from low signal-to-noise ratio (SNR), limiting accuracy to estimate metabolite intensities. A method to coherently sum spectra in a region of interest of the human brain by appropriate peak alignment was developed to yield a mean spectrum with increased SNR. Furthermore, principal component (PC) spectra were calculated to estimate the variance of the mean spectrum. The mean or alternatively the first PC (PC 1 ) spectrum from the same region can be used for quantitation of peak areas of metabolites in the human brain at increased SNR. Monte Carlo simulations showed that both mean and PC 1 spectra were more accurate in estimating regional metabolite concentrations than solutions that regress individual spectra against the tissue compositions of MRSI voxels. Back-to-back MRSI studies on 10 healthy volunteers showed that mean spectra markedly improved reliability of brain metabolite measurements, most notably for myo-inositol, as compared to regression methods. Both single voxel MR spectroscopy (MRS) and MR spectroscopic imaging (MRSI) methods have been used to measure metabolite changes in various neurodegenerative diseases, such as Alzheimer's disease, epilepsy, and amyotrophic lateral sclerosis (1). While MRSI is more efficient in assessing regional distributions of cerebral metabolites than MRS, it is technically more challenging. When using short echo time (TE) acquisitions, resonances from scalp lipids and residual water can severely contaminate metabolite spectra because of "voxel bleeding." Using a combination of lipid nulling and k-space extrapolation (2), we previously achieved substantial reduction of lipid contaminations in proton ( 1 H) MRSI spectra for echo times as short as 25 ms. Although we were able to measure regional distribution of N-acetylaspartate (NAA, a marker of neuronal integrity), choline (Cho), and creatine (Cr) containing compounds with high reliability (3), measurements of myo-inositol (mI, a proposed glial marker) (4) were less reliable (3). We attributed the high variability of mI measurements to a combination of a low signal to noise ratio (SNR) of the mI multiplet peaks and baseline fluctuations resulting from the close spectral proximity of mI and water, in addition to instrumental and biologic variability.To reduce problems from low SNR and baseline fluctuations, we have developed a new approach for data analysis of MRSI. To increase SNR, we averaged spectra from a particular tissue type, e.g., cortical gray matter, within a particular anatomic structure in the brain, e.g., frontal lobe. Aligning the spectral peaks and phasing the data in each individual spectrum before summation maximized the gain in averaged SNR, yielding what we term a regional coherently averaged mean spectrum (RCMS). Under the assumption that baseline fluctuations and instrumental noise are random, an SNR gain of ͌ N can theoretically be achieved for metabolites using RCMS, where N is the number of averaged spectra.We previously demonstrated anot...

MRS signal quantitation: A review of time- and frequency-domain methods

“…More advanced techniques consist of subtracting a modeled macromolecular signal in the frequency domain from the original spectrum. Models may be generated with wavelets [71,132,133,95,75] or splines [134]. A comparison between wavelets and splines has been done in [130] but no significant differences have been found.…”

Test‐retest reliability and reproducibility of short‐echo‐time spectroscopic imaging of human brain at 3T

“…Although a number of past studies have examined the reproducibility or reliability of 1 H‐MRSI (12–20), most have been based on few subjects, some were conducted with nonstandard pulse sequences, and none have been conducted at 3T using a short TE. Furthermore, the majority of these studies were not based on data corrected for partial volume effects, nor has a consistent interscan interval or measure of reproducibility or reliability been applied across studies.…”

“…Furthermore, the majority of these studies were not based on data corrected for partial volume effects, nor has a consistent interscan interval or measure of reproducibility or reliability been applied across studies. Only two studies (13, 16) used a standard measure of test‐retest measurement reliability, the intraclass correlation coefficient (ICC), while two others (12, 14) used an analysis of variance approach to calculate coefficients of variation (CVs) as a measure of reproducibility, dividing the square root of separated variance components (for subject, scan, or voxel) by the mean across all voxels. The remainder of the studies reported more conventional CVs, based on the means and standard deviations of metabolite intensities over similar voxels across subjects or repeated scans.…”

Effects of Alzheimer Disease on Fronto-parietal Brain N-acetyl Aspartate and Myo-Inositol Using Magnetic Resonance Spectroscopic Imaging

“…39,40 In the present study, we used RCMS to determine regional and gray/white matter variations of mI and other metabolites, including NAA, and creatine (Cr) in AD and in cognitively normal (CN) elderly subjects.…”