Preliminary MRI quality assessment and device acceptance guidelines for a multicenter bioclinical study: the GO Glioblastoma Project.

It is a major challenge to guarantee homogeneous acquisition during a prospective multicenter magnetic resonance imaging (MRI) study that makes use of different devices. The goal of the multicenter Grand Ouest Glioblastoma Project (GOGP) was to correlate MRI quantitative parameters with biological markers extracted from image-guided biopsies. Therefore, it was essential to ensure spatial coherence of the parameters as well as the signal intensity and homogeneity. The project included the same MRI protocol implemented on six devices from different manufacturers. The key point was the initial acceptance of the imaging devices and protocol sequences. For this purpose, and to allow comparison of quantitative patient data, we propose a specific method for quality assessment. A common quality control based on 10 parameters was established. Three pulse sequences of the clinical project protocol were applied using three test-objects. A fourth test-object was used to assess T1 accuracy. Although geometry-related parameters, signal-to-noise ratio, uniformity, and T1 measurements varied slightly depending on the different devices, they nevertheless remained within the recommendations and expectations of the multicenter project. This kind of quality control procedure should be undertaken as a prerequisite to any multicenter clinical project involving quantitative MRI and comparison of data acquisitions with quantitative biological image-guided biopsies.

Can dynamic contrast-enhanced magnetic resonance imaging combined with texture analysis differentiate malignant glioneuronal tumors from other glioblastoma?

An interesting approach has been proposed to differentiate malignant glioneuronal tumors (MGNTs) as a subclass of the WHO grade III and IV malignant gliomas. MGNT histologically resemble any WHO grade III or IV glioma but have a different biological behavior, presenting a survival twice longer as WHO glioblastomas and a lower occurrence of metastases. However, neurofilament protein immunostaining was required for identification of MGNT. Using two complementary methods, dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) and texture analysis (MRI-TA) from the same acquisition process, the challenge is to in vivo identify MGNT and demonstrate that MRI postprocessing could contribute to a better typing and grading of glioblastoma. Results are obtained on a preliminary group of 19 patients a posteriori selected for a blind investigation of DCE T1-weighted and TA at 1.5 T. The optimal classification (0/11 misclassified MGNT) is obtained by combining the two methods, DCE-MRI and MRI-TA.

The impact of image dynamic range on texture classification of brain white matter.

BACKGROUND: The Greylevel Cooccurrence Matrix method (COM) is one of the most promising methods used in Texture Analysis of Magnetic Resonance Images. This method provides statistical information about the spatial distribution of greylevels in the image which can be used for classification of different tissue regions. Optimizing the size and complexity of the COM has the potential to enhance the reliability of Texture Analysis results. In this paper we investigate the effect of matrix size and calculation approach on the ability of COM to discriminate between peritumoral white matter and other white matter regions. METHOD: MR images were obtained from patients with histologically confirmed brain glioblastoma using MRI at 3-T giving isotropic resolution of 1 mm3. Three Regions of Interest (ROI) were outlined in visually normal white matter on three image slices based on relative distance from the tumor: one peritumoral white matter region and two distant white matter regions on both hemispheres. Volumes of Interest (VOI) were composed from the three slices. Two different calculation approaches for COM were used: i) Classical approach (CCOM) on each individual ROI, and ii) Three Dimensional approach (3DCOM) calculated on VOIs. For, each calculation approach five dynamic ranges (number of greylevels N) were investigated (N = 16, 32, 64, 128, and 256). RESULTS: Classification showed that peritumoral white matter always represents a homogenous class, separate from other white matter, regardless of the value of N or the calculation approach used. The best test measures (sensitivity and specificity) for average CCOM were obtained for N = 128. These measures were also optimal for 3DCOM with N = 128, which additionally showed a balanced tradeoff between the measures. CONCLUSION: We conclude that the dynamic range used for COM calculation significantly influences the classification results for identical samples. In order to obtain more reliable classification results with COM, the dynamic range must be optimized to avoid too small or sparse matrices. Larger dynamic ranges for COM calculations do not necessarily give better texture results; they might increase the computation costs and limit the method performance.

Texture analysis of magnetic resonance images of rat muscles during atrophy and regeneration.

OBJECTIVES: The goals of the current study were (i) to introduce texture analysis on magnetic resonance imaging (MRI-TA) as a noninvasive method of muscle investigation that can discriminate three muscle conditions in rats; these are normal, atrophy and regeneration; and (ii) to show consistency between MRI-TA results and histological results of muscle type 2 fibers' cross-sectional area. METHOD: Twenty-three adult female Wistar rats were randomized into (i) control (C), (ii) immobilized (I) and (iii) recovering (R) groups. For the last two groups, the right hind limb calf muscles were immobilized against the abdomen for 14 days; then, the hind limb was remobilized only for the R group for 40 days. At the end of each experimental period, MRI was performed using 7-T magnet Bruker Avance DRX 300 (Bruker, Wissembourg); T1-weighted MRI acquisition parameters were applied to show predominantly muscle fibers. Rats were sacrificed, and the gastrocnemius muscle (GM) was excised immediately after imaging. (A) Histology: GM type 2 fibers (fast twitch) were selectively stained using the adenosine triphosphatase (ATPase) technique. The mean cross-sectional areas were compared between the three groups. (B) Image analysis: regions of interest (ROIs) were selected on GM MR images where statistical methods of texture analysis were applied followed by linear discriminant analysis (LDA) and classification. RESULTS: Histological analysis showed that the fibers' mean cross-sectional areas on GM transversal sections represented a significant statistical difference between I and C rats (ANOVA, P<.001) as well as between R and I rats (ANOVA, P<.01), but not between C and R rats. Similarly, MRI-TA on GM transversal images detected different texture for each group with the highest discrimination values (Fisher F coefficient) between the C and I groups, as well as between I and R groups. The lowest discrimination values were found between C and R groups. LDA showed three texture classes schematically separated. CONCLUSION: Quantitative results of MRI-TA were statistically consistent with histology. MRI-TA can be considered as a potentially interesting, reproducible and nondestructive method for muscle examination during atrophy and regeneration.

Proton NMR visible mobile lipid signals in sensitive and multidrug-resistant K562 cells are modulated by rafts.

BACKGROUND: Most cancer cells are characterized by mobile lipids visible on proton NMR (1H-NMR), these being comprised mainly of methyl and methylene signals from lipid acyl chains. Erythroleukemia K562 cells show narrow signals at 1.3 and 0.9 ppm, corresponding to mobile lipids (methylene and methyl, respectively), which are reduced when K562 cells are multidrug resistant (MDR). While the significance of the mobile lipids is unknown, their subcellular localization is still a matter of debate and may lie in the membrane or the cytoplasm. In this study, we investigate the role of cholesterol in the generation of mobile lipid signals. RESULTS: The proportion of esterified cholesterol was found to be higher in K562-sensitive cells than in resistant cells, while the total cholesterol content was identical in both cell lines. Cholesterol extraction in the K562 wild type (K562wt) cell line and its MDR counterpart (K562adr), using methyl-beta-cyclodextrin, was accompanied by a rise of mobile lipids in K562wt cells only. The absence of caveolae was checked by searching for the caveolin-1 protein in K562wt and K562adr cells. However, cholesterol was enriched in another membrane microdomain designated as "detergent-insoluble glycosphingomyelin complexes" or rafts. These microdomains were studied after extraction with triton X-100, a mild non-ionic detergent, revealing mobile lipid signals preserved only in the K562wt spectra. Moreover, following perturbation/disruption of these microdomains using sphingomyelinase, mobile lipids increased only in K562wt cells. CONCLUSION: These results suggest that cholesterol and sphingomyelin are involved in mobile lipid generation via microdomains of detergent-insoluble glycosphingomyelin complexes such as rafts. Increasing our knowledge of membrane microdomains in sensitive and resistant cell lines may open up new possibilities in resistance reversion.

Three dimensional texture analysis in MRI: a preliminary evaluation in gliomas.

The discrimination of tumor boundaries from normal tissue, as well as the evaluation of tissue heterogeneity and tumor grading often continue to pose a challenge in MRI. Although yielding promising results in various fields of medical imaging, two- dimensional (2D) texture analysis in MRI has, until now, demonstrated a lack of specificity in brain tumor classification. A new three-dimensional (3D) approach using Cooccurrence Matrix analysis is proposed to increase the sensitivity and specificity of brain tumor characterization. A preliminary comparative evaluation of 2D and 3D texture analysis was performed on T(1)-weighted MRI of seven gliomas for characterization of solid tumor, necrosis, edema and surrounding white matter. With 3D compared to 2D method, a better discrimination is obtained between necrosis and solid tumor as well as between edema and solid tumor. Using both methods, peritumoral white matter overlaps with edema, but is completely separated from far homo-lateral matter. This latter shows a complete overlapping with contra-lateral matter. The 3D texture analysis approach could provide a new tool for tumor grading and treatment follow-up, as well as for surgery or radiation therapy planning.

Detection of acoustic waves by NMR using a radiofrequency field gradient.

A B(1) field gradient-based method previously described for the detection of mechanical vibrations has been applied to detect oscillatory motions in condensed matter originated from acoustic waves. A ladder-shaped coil generating a quasi-constant RF-field gradient was associated with a motion-encoding NMR sequence consisting in a repetitive binomial 13;31; RF pulse train (stroboscopic acquisition). The NMR response of a gel phantom subject to acoustic wave excitation in the 20-200 Hz range was investigated. Results showed a linear relationship between the NMR signal and the wave amplitude and a spectroscopic selectivity of the NMR sequence with respect to the input acoustic frequency. Spin displacements as short as a few tens of nanometers were able to be detected with this method.