Group independent component analysis of MR spectra

Abstract

This study investigates the potential of independent component analysis (ICA) to provide a data-driven approach for group level analysis of magnetic resonance (MR) spectra. ICA collectively analyzes data to identify maximally independent components, each of which captures covarying resonances, including those from different metabolic sources. A comparative evaluation of the ICA approach with the more established LCModel method in analyzing two different noise-free, artifact-free, simulated data sets of known compositions is presented. The results from such ideal simulations demonstrate the ability of data-driven ICA to decompose data and accurately extract components resembling modeled basis spectra from both data sets, whereas the LCModel results suffer when the underlying model deviates from assumptions, thus highlighting the sensitivity of model-based approaches to modeling inaccuracies. Analyses with simulated data show that independent component weights are good estimates of concentrations, even of metabolites with low intensity singlet peaks, such as scyllo-inositol. ICA is also applied to single voxel spectra from 193 subjects, without correcting for baseline variations, line-width broadening or noise. The results provide evidence that, despite the presence of confounding artifacts, ICA can be used to analyze in vivo spectra and extract resonances of interest. ICA is a promising technique for decomposing MR spectral data into components resembling metabolite resonances, and therefore has the potential to provide a data-driven alternative to the use of metabolite concentrations derived from curve-fitting individual spectra in making group comparisons.

Multimodal Neuroimaging in Schizophrenia: Description and Dissemination

“…Overall these results show high reproducibility and test-retest reliability of tissue-specific estimates of several metabolites using 1 H–MRS at 3 T and short TE in a group of healthy subjects. In addition, a data-driven approach to analyze 1 H–MRS data using ICA was developed (Kalyanam et al 2013 , 2015 ). We are in the process of applying this approach to the spectroscopy data collected from controls and schizophrenia patients.…”

Methodological consensus on clinical proton MRS of the brain: Review and recommendations

“…The relative CRLB derived from fitting each spectrum to a model function is misleading as a numeric estimate of data quality if there is a real absence of specific metabolites; however, good CRLB values may also arise from fitting bad‐quality spectra if the noise is underestimated, artifacts are present, or the fitting method has converged to an incorrect solution (local minimum). Alternative quality measures include CRLB values from a fit to a co‐located water signal, using confidence limits and linewidths from spectral fitting of metabolites or water, detection of outlying values in the spectrum, and use of pattern recognition to classify poor‐quality spectra . A quality map enables easy interpretation at the time of the clinical read, such as implemented in the MIDAS software (Figure ).…”

Application of ICA to realistically simulated 1H‐MRS data

“…Table 2010 shows that both LCModel and ICA improved their ideal data performance compared to our prior results (Kalyanam et al. 2013 ). This is a direct consequence of the recent improvements made in our simulated data generation.…”

“…It is the intent of this study to significantly extend the noise- and artifact-free simulations described above, and also reported in our previous work, to obtain realistic simulated data (Kalyanam et al. 2013 ). To accomplish that, we closely examine the in vivo data, which is an average fid signal from multiple acquisitions.…”

“…In our earlier study we demonstrated the robust performance of ICA in resolving ideal, noise-, and artifact-free-simulated data compared to LCModel, using two different spectral generative models, and observed the need for unambiguous evaluation of in vivo performances of the two methods (Kalyanam et al. 2013 ).…”