Fourier decomposition pulmonary MRI using a variable flip angle balanced steady-state free precession technique.

PURPOSE: Fourier decomposition (FD) is a noninvasive method for assessing ventilation and perfusion-related information in the lungs. However, the technique has a low signal-to-noise ratio (SNR) in the lung parenchyma. We present an approach to increase the SNR in both morphological and functional images. METHODS: The data used to create functional FD images are usually acquired using a standard balanced steady-state free precession (bSSFP) sequence. In the standard sequence, the possible range of the flip angle is restricted due to specific absorption rate (SAR) limitations. Thus, using a variable flip angle approach as an optimization is possible. This was validated using measurements from a phantom and six healthy volunteers. RESULTS: The SNR in both the morphological and functional FD images was increased by 32%, while the SAR restrictions were kept unchanged. Furthermore, due to the higher SNR, the effective resolution of the functional images was increased visibly. The variable flip angle approach did not introduce any new transient artifacts, and blurring artifacts were minimized. CONCLUSION: Both a gain in SNR and an effective resolution gain in functional lung images can be obtained using the FD method in conjunction with a variable flip angle optimized bSSFP sequence.

High temporal versus high spatial resolution in MR quantitative pulmonary perfusion imaging of two-year old children after congenital diaphragmatic hernia repair.

OBJECTIVES: Congenital diaphragmatic hernia (CDH) leads to lung hypoplasia. Using dynamic contrast-enhanced (DCE) MR imaging, lung perfusion can be quantified. As MR perfusion values depend on temporal resolution, we compared two protocols to investigate whether ipsilateral lung perfusion is impaired after CDH, whether there are protocol-dependent differences, and which protocol is preferred. METHODS: DCE-MRI was performed in 36 2-year old children after CDH on a 3 T MRI system; protocol A (n = 18) based on a high spatial (3.0 s; voxel: 1.25 mm(3)) and protocol B (n = 18) on a high temporal resolution (1.5 s; voxel: 2 mm(3)). Pulmonary blood flow (PBF), pulmonary blood volume (PBV), mean transit time (MTT), and peak-contrast-to-noise-ratio (PCNR) were quantified. RESULTS: PBF was reduced ipsilaterally, with ipsilateral PBF of 45 ± 26 ml/100 ml/min to contralateral PBF of 63 ± 28 ml/100 ml/min (p = 0.0016) for protocol A; and for protocol B, side differences were equivalent (ipsilateral PBF = 62 ± 24 vs. contralateral PBF = 85 ± 30 ml/100 ml/min; p = 0.0034). PCNR was higher for protocol B (30 ± 18 vs. 20 ± 9; p = 0.0294). Protocol B showed higher values of PBF in comparison to protocol A (p always <0.05). CONCLUSIONS: Ipsilateral lung perfusion is reduced in 2-year old children following CDH repair. Higher temporal resolution and increased voxel size show a gain in PCNR and lead to higher perfusion values. Protocol B is therefore preferred. KEY POINTS: • Quantitative lung perfusion parameters depend on temporal and spatial resolution. • Reduction of lung perfusion in CDH can be measured with different MR protocols. • Temporal resolution of 1.5 s with spatial resolution of 2 mm (3) is suitable.

In vitro mapping of 1H ultrashort T2* and T2 of porcine menisci.

In this study, mapping of ultrashort T2 and T2* of acutely isolated porcine menisci at B0 = 9.4 T was investigated. Maps of T2 were measured from a slice through the pars intermedia with a spin echo-prepared two-dimensional ultrashort-TE T2 mapping technique published previously. T2* mapping was performed by two-dimensional ultrashort-TE MRI with variable acquisition delay. The measured signal decays were fitted by monoexponential, biexponential and Gaussian-exponential fitting functions. The occurrence of Gaussian-like signal decays is outlined theoretically. The quality of the curve fits was visualized by mapping the value δ = abs(1 - χ(2) red). For T2 mapping, the Gaussian-exponential fit showed the best performance, whereas the monoexponential and biexponential fits showed regionally high values of δ (δ > 20). Interpretation of the Gaussian-exponential parameter maps was found to be difficult, because a Gaussian signal component can be related to mesoscopic (collagen texture) or macroscopic (slice profile, shim, sample geometry) magnetic field inhomogeneities and/or residual (1) H dipole-dipole couplings. It seems likely that an interplay of these effects yielded the observed signal decays. Modulation of the T2* signal decay caused by chemical shift was observed and addressed to fat protons by means of histology. In the T2 measurements, no modulation of the signal decay was observed and the biexponential and Gaussian-exponential fits showed the best performance with comparable values of δ. Our results suggest that T2 mapping provides the more robust method for the characterization of meniscal tissue by means of MRI relaxometry. However, mapping of ultrashort T2, as performed in this study, is time consuming and provides less signal-to-noise ratio per time than the mapping of T2*. If T2* mapping is used, pixel-wise monitoring of the fitting quality based on reduced χ(2) should be employed and great care should be taken when interpreting the parameter maps of the fits.

A novel temporal filtering strategy for functional MRI using UNFOLD.

A major challenge for fMRI at high spatial resolution is the limited temporal resolution. The UNFOLD method increases image acquisition speed and potentially enables high acceleration factors in fMRI. Spatial aliasing artifacts due to interleaved k-space sampling are to be removed from the image time series by temporal filtering before statistical mapping in the time domain can be carried out. So far, low-pass filtering and multi-band filtering have been proposed. Particularly at high UNFOLD factors both methods are non-optimal. Low-pass filtering severely degrades temporal resolution and multi-band filtering leads to temporal autocorrelations affecting statistical modelling of activation. In this work, we present a novel temporal filtering strategy that significantly reduces temporal autocorrelations compared to multi-band filtering. Two datasets (finger-tapping and resting state) were post-processed using the proposed and the multi-band filter with varying set-ups (i.e. transition bands). When the proposed filtering strategy was used, a linear regression analysis revealed that the number of false positives was significantly decreased up to 34% whereas the number of activated voxels was not significantly affected for most filter parameters. In total, this led to an effective increase in the number of activated voxels per false positive for each filter set-up. At a significance level of 5%, the number of activated voxels was increased up to 41% by using the proposed filtering strategy.

Assessment of renal function after conformal radiotherapy and intensity-modulated radiotherapy by functional 1H-MRI and 23Na-MRI.

PURPOSE: Adjuvant radiochemotherapy (RCHT) improves survival of patients with locally advanced gastric cancer. Conventional three-dimensional conformal radiotherapy (3D-CRT) results in ablative doses to a significant amount of the left kidney, while image-guided intensity-modulated radiotherapy (IG-IMRT) provides excellent target coverage with improved kidney sparing. Few long-term results on IMRT for gastric cancer, however, have been published. Functional magnetic resonance imaging (fMRI) at 3.0 T including blood oxygenation-level dependent (BOLD) imaging, diffusion-weighted imaging (DWI) and, for the first time, (23)Na imaging was used to evaluate renal status after radiotherapy with 3D-CRT or IG-IMRT. PATIENTS AND METHODS: Four disease-free patients (2 after 3D-CRT and 2 after IMRT; FU for all patients > 5 years) were included in this feasibility study. Morphological sequences, axial DWI images, 2D-gradient echo (GRE)-BOLD images, and (23)Na images were acquired. Mean values/standard deviations for ((23)Na), the apparent diffusion coefficient (ADC), and R2* values were calculated for the upper/middle/lower parts of both kidneys. Corticomedullary (23)Na-concentration gradients were determined. RESULTS: Surprisingly, IG-IMRT patients showed no morphological alterations and no statistically significant differences of ADC and R2* values in all renal parts. Values for mean corticomedullary (23)Na-concentration matched those for healthy volunteers. Results were similar in 3D-CRT patients, except for the cranial part of the left kidney. This was atrophic and presented significantly reduced functional parameters (p = 0.001-p = 0.033). Reduced ADC values indicated reduced cell density and reduced extracellular space. Cortical and medullary R2* values of the left cranial kidney in the 3D-CRT group were higher, indicating more deoxygenated hemoglobin due to reduced blood flow/oxygenation. ((23)Na) of the renal cranial parts in the 3D-CRT group was significantly reduced, while the expected corticomedullary (23)Na-concentration gradient was partially conserved. CONCLUSIONS: Functional MRI can assess postradiotherapeutic renal changes. As expected, marked morphological/functional effects were observed in high-dose areas (3D-CRT), while, unexpectedly, no alteration in kidney function was observed in IG-IMRT patients, supporting the hypothesis that reducing total/fractional dose to the renal parenchyma by IMRT is clinically beneficial.

Quantitative pulmonary perfusion imaging at 3.0 T of 2-year-old children after congenital diaphragmatic hernia repair: initial results.

OBJECTIVE: To investigate whether dynamic contrast-enhanced MR imaging of the lung following congenital diaphragmatic hernia repair is feasible at 3.0 T in 2-year-old children and whether associated lung hypoplasia (reflected in reduced pulmonary microcirculation) can be demonstrated in MRI. METHODS: Twelve children with a mean age 2.0 ± 0.2 years after hernia repair underwent DCE-MRI at 3.0 T using a time-resolved angiography with stochastic trajectories sequence. Quantification of lung perfusion was performed using a pixel-by-pixel deconvolution approach. Six regions of interest were placed (upper, middle and lower parts of right and left lung) to assess differences in pulmonary blood flow (PBF), pulmonary blood volume (PBV) and mean transit time (MTT) while avoiding the inclusion of larger pulmonary arteries and veins. RESULTS: The difference in PBF and PBV between ipsilateral and contralateral lung was significant (P < 0.5). No significant differences could be detected for the MTT (P = 0.5). CONCLUSION: DCE-MRI in 2-year-old patients is feasible at 3.0 T. Reduced perfusion in the ipsilateral lung is reflected by significantly lower PBF values compared with the contralateral lung. DCE-MRI of the lung in congenital diaphragmatic hernia can help to characterise lung hypoplasia initially and in the long-term follow-up of children after diaphragmatic repair. KEY POINTS: Congenital diaphragmatic hernia often leads to lung hypoplasia and secondary pulmonary hypertension. Dynamic contrast-enhanced 3-T magnetic resonance can assess these complications in 2-year-olds. The affected ipsilateral lung shows reduced perfusion and lower pulmonary blood flow. Thoracic DCE-MRI helps characterise lung hypoplasia in children after hernia repair.

Radiotherapy treatment planning of basal meningiomas: improved tumor localization by correlation of CT and MR imaging data.

A localization technique, based on three-dimensional CT and MR imaging data for precision radiotherapy of basal meningiomas, is presented. Indications for radiotherapy included unresected tumors, gross disease remaining despite surgery, and recurrences. The patient's head was fixed in a stereotactic localization system which is usable at the CT, MR and the linear accelerator installations. The geometrical distortion of MR imaging data was evaluated in three dimensions by phantom measurements. The geometrical distortion was "corrected" (reducing displacements to the size of a pixel) by calculations based on modelling the distortion as a fourth order two-dimensional polynomial. The target volume was defined in three-dimensional MR imaging data after application of 0.1 mmol/kg b.w. Gd-DTPA solution and transferred precisely from MR onto CT data to provide a map of the radiation attenuation coefficient for dose calculation. The superior soft tissue contrast of MR showed an excellent tumor delineation especially when the bony base of the skull obscured the target in CT images. Target volume, calculated dose distribution, and critical structures could be transferred between CT and MR imaging data and displayed as three-dimensional shaded structures for better assessment for matching of target volume and dose distribution. With the described planning system a more precise target definition of basal meningiomas was possible by integration of the superior tumor delineation in MR compared with CT.

Improved target volume definition for precision radiotherapy planning of meningiomas by correlation of CT and dynamic, Gd-DTPA-enhanced FLASH MR imaging.

In this methodological paper the authors report a fast, T1-weighted gradient-echo sequence (FLASH) for dynamic, Gd-DTPA-enhanced magnetic resonance (MR) imaging of meningiomas and its application in precision radiotherapy planning. Indications for radiotherapy included unresected tumors, tumor remaining after surgery, and recurrences. The patient's head was fixed in a stereotactic localization system which is usable at the CT, MR and the linear accelerator installations. By phantom measurements different materials (steel, aluminum, titanium, plastic, wood, ceramics) used for the stereotactic system were tested for mechanical stability and geometric MR image distortion. All metallic stereotactic rings (closed rings made of massive metal) led to a more or less dramatic geometric distortion and signal cancellation in the MR images. The best properties--nearly no distortion and high mechanic stability--are provided by a ceramic ring. If necessary, the remaining geometric MR image distortion can be 'corrected' (reducing displacements to the size of a pixel) by calculations based on modeling the distortion as a fourth order two-dimensional polynomial. The target volume was defined in dynamic, T1-weighted FLASH MR images, which were measured before, during, and after the controlled intravenous infusion of 0.1 mmol/kg body weight Gd-DTPA. The stereotactic localization technique allows the precise transfer of the target volume information from MR onto CT data to provide a map of the radiation attenuation coefficient for dose calculation. In genera, the superior soft tissue contrast of MR showed an excellent tumor delineation, especially in regions, such as the base of the skull, where the target often was obscured in CT images.(ABSTRACT TRUNCATED AT 250 WORDS)

Corpus callosum in Alzheimer's disease and vascular dementia--a quantitative magnetic resonance study.

We investigated atrophic alterations in different regions of the corpus callosum in Alzheimer's disease (AD) and vascular dementia (VD) with respect to clinical changes. 32 patients with AD (NINCDS-ADRDA criteria), 17 patients with VD (NINDS-AIREN criteria) and 13 healthy control subjects were included. 3-D MRI sequences were acquired using a 1.5T MRI scanner. The size of the corpus callosum and its subdivisions was sampled on 5 mid-saggital slices using a personal computer-based software. Total callosal size was significantly reduced in AD but not in VD. Furthermore, the most rostral parts of the corpus callosum were significantly smaller in AD when compared to controls. Again, these changes were not found in patients with VD. Severity of dementia was significantly correlated with the size of the midbody of the corpus callosum in AD. Callosal atrophy in AD may reflect the severity and pattern of cortical neuronal damage occurring mostly in the inferior frontal, anterior parietal and midtemporal regions. Correlations between regional callosal atrophy and severity of dementia indicate that interhemispheric cortico-cortical disconnections may contribute to the dementia syndrome.

Quantitative magnetic resonance imaging in geriatric depression and primary degenerative dementia.

Quantitative magnetic resonance imaging (MRI) was used to investigate volumes of different brain structures in 19 patients with late-onset major depression (DSM-III-R), 27 patients with Alzheimer's disease (NINCDS-ADRDA criteria) and 13 age matched controls. 3-D MRI sequences were acquired using a Siemens 1.5 T scanner. Whole brain volume, CSF volume, volume of the frontal and temporal lobes and the volume of the amygdala-hippocampus complex were assessed using the software NMR Win. Compared to the controls, depressed patients showed a significantly lower whole brain volume and a significantly higher CSF volume, whereas volumes of the frontal and temporal lobes as well as the amygdala-hippocampus complex volumes were not significantly decreased. In addition, depressed patients exhibited a higher ventricle-brain ratio suggesting a higher degree of central atrophy compared to healthy individuals. In contrast, Alzheimer patients showed significantly lower volumes than depressed patients and controls with respect to all volumetric parameters. Although the findings indicate the presence of brain atrophy in patients with late-onset depression, the pattern of volumetric changes in these patients differs markedly from that observed in patients with primary degenerative dementia.

Evaluation of proton density by magnetic resonance imaging: phantom experiments and analysis of multiple component proton transverse relaxation.

The quantitative evaluation of proton density by magnetic resonance imaging (MRI) is limited as a result of non-uniformities in the intensity distribution of the images and by the fact that only part of the protons of the tissue contribute to the image signal. This study was undertaken to estimate the accuracy of proton density measurements using a standard whole-body MR imager operating at 1.5 T. First, phantom experiments were performed to examine the possibility of an intensity correction. For the test phantom the systematical errors in the computed proton densities were reduced from 5 to 1% after correction. Secondly, proton transverse relaxation curves of biological tissues were measured in vitro on an MR spectrometer. A multi-exponential analysis of the data shows that for spin-echo times TE greater than 10 ms in total between 10 and 30% of the protons of the tissue do not contribute to the image signal. In all tissues a proton component with a free induction decay (FID) time T2* less than 32 microseconds was observed. In the time range TE greater than 10 ms two proton components can be distinguished in muscle and fatty tissue. Finally, it will be shown that a pixel-orientated two-exponential analysis of spin-echo images leads to a much more homogeneous density image than one-exponential computation, since tissue-specific biexponentiality and partial volume effects are taken into account. As a conclusion, the hydrogen density of biological tissues can be evaluated at best with an overall error of 10% from MR images for TE greater than 10 ms. This accuracy is insufficient for a pixel-orientated neutron therapy planning.

Estimation of aortic compliance using magnetic resonance pulse wave velocity measurement.

A method for compliance estimation employing magnetic resonance pulse wave velocity measurement is presented. Time-resolved flow waves are recorded at several positions along the vessel using a phase contrast sequence, and pulse wave velocity is calculated from the delay of the wave onsets. Using retrospective cardiac gating in combination with an optically decoupled electrocardiogram acquisition, a high temporal resolution of 3 ms can be achieved. A phantom set-up for the simulation of pulsatile flow in a compliant vessel is described. In the phantom, relative errors of pulse wave velocity estimation were found to be about 15%, whereas in a volunteer, larger errors were found that might be caused by vessel branches. Results of pulse wave velocity estimation agree with direct aortic distension measurements which rely on a peripheral estimate of aortic pressure and are therefore less accurate. Studies in 12 volunteers show values of pulse wave velocity consistent with the literature; in particular the well-known increase in pulse wave velocity with age was observed. Preliminary results show that the method can be applied to aortic aneurysms.

Three-dimensional BANG gel dosimetry in conformal carbon ion radiotherapy.

In this study we applied BANG polymer-gel dosimetry using magnetic resonance imaging (MRI) to densely ionizing radiation such as carbon ion beams. BANG polymer gels were irradiated with a quadratic field of monoenergetic 12C ions at different beam energies in the range of 135 MeV u(-1) to 410 MeV u(-1). They were irradiated at the radiotherapy facility of the GSI, Darmstadt, Germany. Our object was to examine the saturation effect for densely ionizing radiation that occurs at high values of linear energy transfer (LET). The examination yielded the first effectiveness values that will be discussed in the following sections. A solid sphere and a hollow sphere were both irradiated with a horizontal pencil beam from the raster scanning facility at energies of 268 MeV u(-1) (solid sphere) and 304 MeV u(-1) (hollow sphere) respectively. MR dosimetry measurements were compared with data from a planning system. As far as quality is concerned, there is good agreement between the measured dose distributions of both samples and the dose maps from the planning software. The measured MR signals cannot be converted into absolute dose, since the relative efficiency is still unknown for mixed radiation fields of primary carbon ions and it is known only to a limited extent for nuclear fragments with different energies from highly energetic photon radiation. Model calculations are in progress in order to facilitate conversions of measured MR signals into dose.

Topography of callosal atrophy reflects distribution of regional cerebral volume reduction in Alzheimer's disease.

It has been suggested that regional corpus callosum atrophy in Alzheimer's disease (AD) may serve as an in vivo index of neuronal loss in the neocortex. In this study total and regional size of the corpus callosum was evaluated with respect to the volumes of the frontal, temporal, and parietal lobes in 38 patients with AD (NINCDS-ADRDA criteria) using quantitative magnetic resonance imaging. Twenty healthy subjects matched for age and gender served as a control group. All quantitative measurements were performed by manual tracing using personal computer-based software. Both total size and the five measured regional subsections were significantly smaller in AD when compared to the control subjects. The severity of dementia was significantly correlated with the size of the middle sections of the corpus callosum (rostral body and midbody). Within the AD group, the rostral body of the corpus callosum was significantly correlated with the frontal lobe volumes, the midbody was correlated with the temporal lobe volumes, and size of the splenium was correlated with the parietal lobe volumes. We conclude that callosal atrophy in AD reflects the severity and pattern of cortical neuronal damage. Correlations between regional callosal atrophy and severity of dementia indicate that interhemispheric cortico-cortical disconnection may contribute to the dementia syndrome.

The use of reticulated foam in texture test objects for magnetic resonance imaging.

The use of texture analysis in magnetic resonance imaging requires the availability of texture test objects for use in standardisation of in vivo measurement. A series of such objects of varying texture has been developed using reticulated foam, which has an open pore structure. The texture properties of these foams have been compared with in vivo brain and the ability of texture analysis to discriminate the different porosities of foam and brain (white matter) demonstrated.

Comparison of TSE, TGSE, and CPMG measurement techniques for MR polymer gel dosimetry.

A radiation dose distribution that optimally conforms to the target volume is of major interest for stereotactic radiotherapy. For this purpose treatment plans have to be verified experimentally before transferring to the patient. The requirements regarding dose accuracy and spatial resolution can be fulfilled with tissue equivalent polymer gel dosimeters which offer the possibility to visualize 3D dose distributions. Herewith, dosimetry can be performed by the spin-spin relaxation rate R2 which varies with the absorbed dose. In this work, different MR measurement techniques were evaluated: The standard Carr-Purcell-Meiboom-Gill (CPMG) method, a modified Turbo-Spin-Echo (TSE) sequence, and a modified Turbo-Gradient-Spin-Echo (TGSE) sequence. Experiments were performed both with a homogeneous water phantom and an irradiated polymer gel. The results show that TGSE and especially TSE are suited well for MR polymer gel dosimetry: The acquisition time of both techniques can be reduced in comparison to CPMG by a factor of 5. The accuracy of dose determination for doses between 2 Gy and 13 Gy lies between 5.6% and 2.0% (TSE), 9.0% and 3.2% (TGSE), and 7.9% and 2.7% (CPMG). These investigations show that especially TSE can be handled as a substitute or at least an alternative to CPMG for the verification of treatment plans in stereotactic radiotherapy.

1H-spectroscopic imaging using a modified Dixon method.

Inhomogeneities of the static magnetic field and the different susceptibilities of the various types of tissue are a serious problem for all imaging methods of spectral separation of fat and water. In the Dixon method this problem is solved by using the absolute values of the image signals for the separation. In image regions where the fat signal is greater than the water signal, however, this results in an incorrect assignment of the computed solutions. A modified Dixon method was developed to easily carry out the spectral separation completely over the entire image by interactively building up a phase correction matrix after the data acquisition. The spectral delineation of the fat tissue finds an interesting application in the treatment planning with fast neutrons in accounting for the increase in dose.

MR tissue characterization of intracranial tumors by means of texture analysis.

Tissue characterization of the human brain has been performed by texture analysis of proton relaxation time images using a standard MR whole body imager operating at 1.5 T. A combined CP/CPMG multi-echo, multislice sequence was used to measure T1 and T2 in each pixel with an uncertainty not exceeding 10%. In a prospective clinical study, 12 patients with histologically confirmed brain tumors were investigated. For each ROI in the calculated T1 and T2 parameter images, texture parameters originating from the grey level distribution, the gradient distribution, the grey level co-occurrence matrix, and the grey level runlength histogram were used for classification and discrimination between tissues. All regions corresponding to the normal brain tissue (white matter, grey matter, cerebrospinal fluid) were successfully discriminated from each other as well as from the pathological tissue parts (edema and tumor). The classification of 10 edematous and 8 tumorous tissue regions yielded only one misclassification. Together with additional rules, these discrimination rules formed the knowledge base of an expert system for segmentation of the brain images. In cases of tumors without Gd-DTPA contrast medium uptake or in cases of Gd-DTPA contraindication, segmentated images can help solve nontrivial diagnostical problems such as delineating the target volume in radiation therapy.

An amplitude optimized single-shot hybrid QUEST technique.

Rapid MR imaging techniques either deposit high amounts of radio frequency power or require powerful gradient systems with high slew rates, which might not be available on conventional scanners. QUEST provides a fast imaging method with scan times of the order of hundreds of milliseconds and avoids these problems at the cost of low signal-to-noise ratios (SNR). In this work, QUEST was optimized with regard to image quality and measuring time. With the use of a Hybrid QUEST technique, that refocuses the image echoes several times, a spatial resolution of 1.9 mm x 1.6 mm x 5 mm was achieved. By acquiring both the necessary correction data and the image information in a single echo train, the Hybrid QUEST technique was implemented as a true single-shot measurement with a total scan time of 190 ms. Optimization of the excitation flip angles and the amplitude and phase correction methods for image reconstruction resulted in an improved SNR of 53.7 in the white matter of the human head for a 10 mm slice thickness at 1.5 T. In contrast to echo planar imaging techniques, no image distortions were observed with Hybrid QUEST in anatomic regions with many tissue interfaces.

Measurement of diffusion coefficients using a quick echo split NMR imaging technique.

A new method for the ultrafast generation of diffusion-weighted images is reported. The technique combines a quick echo split NMR imaging sequence with the principle of Stejskal and Tanner. It allows to determine the diffusion constant with nearly the same accuracy as the conventional spin-echo technique, requiring only a fraction of the time. The determined values for water doped with 1 g Cu(NO3)2 per liter of H2O and pure acetone were Dwater = (1.95 +/- 0.02) x 10(-9) m2/s and Dacetone = (4.05 +/- 0.02) x 10(-9) m2/s at 18.5 degrees C. They are in good agreement both with literature and our own reference measurements using a diffusion-weighted spin-echo sequence. In addition, the temperature dependence of Dwater was measured in the range of 18.5-45.9 degrees C and a good correspondence with reported data was found.