Spatial and Temporal Resolution

The achievable quality of a spectrum identified by its SNR is directly related to the size of the sample volume (VOI) and the acquisition time, defined by the number of signal averages N multiplied by the repetition time (TR), as described in the paragraph above. Thus the SNR is usually the limiting factor for spatial and temporal resolution in a clinical setting, where the time of acquisition is usually restricted by logistical, technical, financial and ethical aspects.

Spectra can be acquired from a single-voxel or a multidimensional grid of spectra, which are generally referred to as chemical shift or spectroscopic images (CSI, SI) (Fig. 6.4). Unlike the ca. 1 mm3 resolution that can be achieved with routine MRI, 1H-MRS studies have been performed with a spatial resolution of 8 ml and an acquisition time of 3-5 min per single-voxel acquisition and 2-4 ml for CSI in 5-10 min [23].

The greater ability to detect smaller lesions afforded by advanced MRI technologies can involve an increased need for greater spatial resolution of single-voxel spectroscopy and CSI experiments. Whereas the poor SNR of 1.5 T magnets prevented the acquisition of single-voxel spectra with a spatial resolution significantly below 4-8 ml or chemical shift images with a spatial resolution of less than 2 ml, lower spatial resolutions are becoming feasible at 3 T.

The direct matching of metabolic and anatomical information would obviously be of clinical importance since it could improve the sensitivity and specificity of diagnosis. Higher-resolution SI can enable to distinguish between different anatomical structures, between normal and pathological tissue, or between different pathological structures (e.g. heterogeneous tumours, SM plaques, etc.). In patients with brain tumour, for example, high-resolution SI could differentiate healthy oedematous from affected tissue as well as normal from normal-appearing hippocampus in patients with temporal lobe epilepsy. At 3 T and higher field intensities, voxel volumes of 1 cm3 and lower can be obtained with still good intrinsic SNR and acceptable acquisition times [23, 24].

Higher field strengths have also considerably ameliorated resolution time reducing examination times which had limited the application of 1H-MRS to clinical research. The advantages of shorter examinations for patients, radiologists, technicians and hospital administrators are obvious. At high magnetic fields, larger brain volumes can be studied over times similar to current single-voxel protocols using multivoxel 2D or 3D 1H-SI sequences. In multivoxel 3D spectroscopy ex-

Fig. 6.4. Whereas a single-voxel spectrum usually provides good quality, high SNR data but limited to a small region, a CSI acquisitionwillyieldagridofspectra, which areoftenproneto artefacts orlowSNR, which canbetranslatedinto metabolite maps. (Images courtesy of Charité, Virchow Clinic, Berlin, Germany)

Fig. 6.4. Whereas a single-voxel spectrum usually provides good quality, high SNR data but limited to a small region, a CSI acquisitionwillyieldagridofspectra, which areoftenproneto artefacts orlowSNR, which canbetranslatedinto metabolite maps. (Images courtesy of Charité, Virchow Clinic, Berlin, Germany)

periments, the time of acquisition has been reduced 25%, retaining the lowest SNR of 1.5 T at 3T [18, 22, 23]. Shorter times allow multinuclear MRS to determine the metabolic changes coupled to neuronal activity.

Beside the clinical advantages, high spatial and temporal resolution also have some striking technical features. Next to the SNR, the linewidths of the individual peaks are responsible for the qualitative appearance of a spectrum. The linewidth is defined by the T2 relaxation time of the metabolite and the local field inhomo-geneities, where the inhomogeneities can be the dominating factor of a resulting T2*. As local inhomogenei-ties will decrease with smaller voxel volumes, T2* increases, resulting in a noticeable decrease of linewidths, improving spectral quality, especially for voxels with a volume smaller than 0.75 cm3 [24].

Another factor degrading spectral quality is the stability of the consecutively acquired signal averages. Patient motion, magnet drifts and the instability of other system components can interfere with the acquisition of MRS data, yielding broadened metabolite peaks or other artefacts. Thus keeping the acquisition times as short as possible spectral quality can significantly be improved.

Unraveling Alzheimers Disease

Unraveling Alzheimers Disease

I leave absolutely nothing out! Everything that I learned about Alzheimer’s I share with you. This is the most comprehensive report on Alzheimer’s you will ever read. No stone is left unturned in this comprehensive report.

Get My Free Ebook


Post a comment