Ghosting artifacts in spectroscopy are problematic, as they superimpose with metabolites and lead to inaccurate quantification. Detection and removal of ghosting artifacts using traditional machine learning approaches with feature extraction/selection is difficult, as ghosts appear at different frequencies. Here, we show that DL methods perform extremely well for ghost detection if the spectra are treated as images in the form of time-frequency representations. Further optimization for in vivo spectra will hopefully confirm their "ghostbusting" capacity. Magn Reson Med 80:851-863, 2018. © 2018 International Society for Magnetic Resonance in Medicine.

Using the simultaneously acquired water signal as internal reference allows not only for compensation of phase and frequency fluctuations but also for signal amplitude restoration, and thus improved metabolite ADC estimation. Combination with 2D simultaneous fitting promises access to the diffusion properties even for low signal-to-noise metabolites. The combination of both techniques increases the specificity and sensitivity of estimated metabolite ADC values in the cohort.

It is shown that inclusion of a measured lineshape into modeling of interrelated MR spectra is beneficial and can be combined also with simultaneous spectral and diffusion modeling.

To demonstrate that oscillating gradient spin-echo sequences can be combined with diffusion-weighted magnetic resonance spectroscopy even on clinical MR systems to study human brain at short diffusion times to provide apparent diffusion coefficients (ADCs) sensitive to a narrower cellular length scale than pulsed gradient spin-echo sequences at long diffusion time. Methods Measurements were performed on a 3T MR system using a semiLaser sequence with diffusion-weighting realized by oscillating and pulsed gradient modules, encoding diffusion times <10 ms and >50 ms, respectively. Metabolite-cycling was included to measure metabolites and water simultaneously. The sequence was tested in a phantom and in a parietooccipital cerebral region of interest with mixed gray/white matter content of 6 subjects. The water reference was used for phase, frequency, and eddy-current correction as well as motion compensation. ADCs were estimated by 1D sequential and 2D simultaneous fitting. Results Measurements in the phantom established that both sequences yield equal ADCs, independent of diffusion time, as expected for free diffusion. In contrast, on average over multiple metabolites in vivo metabolite diffusion was found to be 1.9 times faster at short (8.3 ms) than at long (155 ms) diffusion times. The difference in ADC was found to be statistically significant for the creatines, cholines, N-acetylaspartate, N-acetylaspartylglutamate, myo-inositol, scyllo-inositol, glutamate, glutamine and taurine. The water ADC was measured to be 1.3 times larger at short than at long diffusion time. Conclusion It is demonstrated that application of oscillating gradients in diffusion-weighted MRS is feasible on clinical MR systems to establish the dependence of ADCs on diffusion times in humans. The initial results largely confirm earlier reports for mice' and rats' brain at short and long diffusion times. ADCs representing diffusion at short and ultra-short diffusion times are of interest to probe cellular or subcellular changes in disease. The presented methodology may thus open the door for investigation of pathophysiological changes in cell-specific microcellular structures in human cohorts.

Purpose: Definition of a macromolecular MR spectrum based on diffusion properties rather than relaxation time differences and characterization of non-Gaussian diffusion of brain metabolites with strongly diffusion-weighted MR spectroscopy. Methods: Short echo time MRS with strong diffusion-weighting with b-valuesup to 25 ms/μm 2 at two diffusion times was implemented on a Connectom system and applied in combination with simultaneous spectral and diffusion decay modeling. Motion-compensation was performed with a combined method based on the simultaneously acquired water and a macromolecular signal. Results: The motion compensation scheme prevented spurious signal decay reflected in very small apparent diffusion constants for macromolecular signal.Macromolecular background signal patterns were determined using multiple fit strategies. Signal decay corresponding to non-Gaussian metabolite diffusion was represented by biexponential fit models yielding parameter estimates for human gray matter that are in line with published rodent data. The optimal fit strategies used constraints for the signal decay of metabolites with limited signal contributions to the overall spectrum. Conclusion:The determined macromolecular spectrum based on diffusion properties deviates from the conventional one derived from longitudinal relaxation time differences calling for further investigation before use as experimental basis spectrum when fitting clinical MR spectra. The biexponential characterization of metabolite signal decay is the basis for investigations into pathologic alterations of microstructure.