1H Magnetic Resonance Spectroscopic Imaging Research Articles

3D characterization of gas phase reactors with regularly and irregularly structured monolithic catalysts by NMR imaging and modeling

A heterogeneously catalyzed gas phase reaction process was characterized regarding temperature and concentration profiles by means of three dimensional (3D) 1H magnetic resonance spectroscopic imaging (MRSI), using the exothermal ethylene hydrogenation as an example. Here, temperature mapping was achieved by using specifically designed thermometers filled with ethylene glycol.The impact of heat and mass transfer on the process performance was investigated by using two different monolithic catalysts with completely different heat and mass transfer characteristics: a regularly structured honeycomb monolith and a irregularly structured open-cell foam packing. The influence of these characteristics on the reaction zones within the monolithic catalysts was demonstrated by simulations that were based on 2D reactor models.To evaluate the applicability of temperature and concentration mapping by 1H MRSI for model validation, a predictive two dimensional model of the process was applied. The resulting simulations of temperature profiles and concentration distributions were in very good agreement with the experimental data with deviations below 9%. Conventional mass spectroscopic measurements provided further evidence of the accuracy of 3D MRSI measurements as well as the 2D reactor model.

Metabolic changes in normal appearing white matter in multiple sclerosis patients using multivoxel magnetic resonance spectroscopy imaging.

Demyelination and axonal degeneration caused by multiple sclerosis (MS) exist in the white matter and not only in the lesion area. Magnetic resonance spectroscopy (MRS) could provide a unique insight into metabolic changes in the normal appearing white matter (NAWM). To evaluate the subtle axonal degeneration and delineate the spatial distribution of metabolite abnormalities in the NAWM in patients with MS. A total of 17 clinically definite relapsing-remitting MS (RRMS) patients and 21 healthy controls were enrolled in this study. 2D 1H magnetic resonance spectroscopic imaging (MRSI) performed at 3 Tesla was used to measure metabolite concentrations in the frontal-parietal-occipital NAWM. Ratios of N-acetyl-aspartate (NAA) and choline (Cho) to creatine (Cr) and Cho to NAA were calculated in each voxel. MS patients showed decreased NAA/Cr and increased Cho/NAA ratios in the NAWM compared to healthy controls. In the parietal NAWM, the extent of NAA/Cr decrease was significantly higher than that in the frontal and parietal-occipital NAWM. Decreased NAA in the NAWM would provide useful metabolic information for evaluation of disease progression in MS. The high extent of NAA decrease in the parietal NAWM helps improve the accuracy of the prediction.

Quantitative multi-modal MR imaging as a non-invasive prognostic tool for patients with recurrent low-grade glioma.

Low-grade gliomas can vary widely in disease course and therefore patient outcome. While current characterization relies on both histological and molecular analysis of tissue resected during surgery, there remains high variability within glioma subtypes in terms of response to treatment and outcome. In this study we hypothesized that parameters obtained from magnetic resonance data would be associated with progression-free survival for patients with recurrent low-grade glioma. The values considered were derived from the analysis of anatomic imaging, diffusion weighted imaging, and 1H magnetic resonance spectroscopic imaging data. Metrics obtained from diffusion and spectroscopic imaging presented strong prognostic capability within the entire population as well as when restricted to astrocytomas, but demonstrated more limited efficacy in the oligodendrogliomas. The results indicate that multi-parametric imaging data may be applied as a non-invasive means of assessing prognosis and may contribute to developing personalized treatment plans for patients with recurrent low-grade glioma.

Association of early changes in 1H MRSI parameters with survival for patients with newly diagnosed glioblastoma receiving a multimodality treatment regimen.

The heterogeneous biology of glioblastoma (GBM) emphasizes the need for imaging methods to assess tumor burden and assist in evaluating individual patients. The purpose of this study was to investigate early changes in metrics from 3D 1H magnetic resonance spectroscopic imaging (MRSI) data, compare them with anatomic lesion volumes, and determine whether they were associated with survival for patients with newly diagnosed GBM receiving a multimodality treatment regimen. Serial MRI and MRSI scans provided estimates of anatomic lesion volumes and levels of choline, creatine, N-acetylaspartate, lactate, and lipid. The association of metrics derived from these data with survival was assessed using Cox proportional hazards models with adjustments for age, Karnofsky performance score, and extent of resection. Temporal changes in parameters were evaluated using a Wilcoxon signed rank test. Anatomic lesion volumes at the post-radiotherapy (RT) scan, metabolic lesion volume at mid-RT and post-RT scans, as well as metrics describing levels of choline, lactate, and lipid were associated with overall survival. There was a significant reduction in the enhancing lesion volume, increase in T2 lesion volume from mid-RT to post-RT, and decrease in parameters describing metabolite levels during these early time points. The MRSI data provided metrics that described the effects of treatment on the metabolic lesion burden and were associated with overall survival. This suggests that adding these parameters to standard assessments of changes in anatomic lesion volumes could contribute to making early decisions about the efficacy of such combination therapies.

Whole-brain patterns of (1)H-magnetic resonance spectroscopy imaging in Alzheimer's disease and dementia with Lewy bodies.

Magnetic resonance spectroscopy has demonstrated metabolite changes in neurodegenerative disorders such as Alzheimer's disease (AD) and dementia with Lewy bodies (DLB); however, their pattern and relationship to clinical symptoms is unclear. To determine whether the spatial patterns of brain-metabolite changes in AD and DLB are regional or diffused, and to examine whether the key metabolite levels are associated with cognitive and non-cognitive symptoms, we acquired whole-brain spatially resolved 3T magnetic resonance spectroscopic imaging (MRSI) data from subjects with AD (N=36), DLB (N=35) and similarly aged controls (N=35). Voxel-wise measurement of N-acetylaspartate to creatine (NAA/Cr), choline to Cr (Cho/Cr), myo-inositol to Cr (mI/Cr) as well as glutamate and glutamine to Cr (Glx/Cr) ratios were determined using MRSI. Compared with controls, AD and DLB groups showed a significant decrease in most brain metabolites, with NAA/Cr, Cho/Cr and mI/Cr levels being reduced in posterior cingulate, thalamus, frontotemporal areas and basal ganglia. The Glx/Cr level was more widely decreased in DLB (posterior cingulate, hippocampus, temporal regions and caudate) than in AD (only in posterior cingulate). DLB was also associated with increased levels of Cho/Cr, NAA/Cr and mI/Cr in occipital regions. Changes in metabolism in the brain were correlated with cognitive and non-cognitive symptoms in the DLB but not in the AD group. The different patterns between AD and DLB may have implications for improving diagnosis, better understanding disease-specific neurobiology and targeting therapeutics. In addition, the study raised important questions about the role of occipital neuroinflammation and glial activation as well as the glutamatergic treatment in DLB.

Impact of an endorectal coil for 1H-magnetic resonance spectroscopy of the prostate at 3.0 T in comparison to 1.5 T: Do we need an endorectal coil?

ObjectivesTo evaluate the influence of endorectal coil (ERC) regarding spectral quality and diagnostic suitability and diagnostic performance in 3.0T 1H-magnetic resonance spectroscopy imaging (MRSI) compared to 1.5T MRSI. Materials and methodsThe study was approved by the Institutional Review Board. MRSI of the prostate was performed on 19 patients at 1.5T with ERC (protocol 1), at 3.0T with a disabled ERC (protocol 2) and at 3.0T with ERC (protocol 3). Age, weight, body size, body-mass-index, prostate volume, time between measurements, diagnostic suitability of spectra, histopathological results after biopsy of cancer suspect lesions (CSL), sensitivity and specificity were evaluated. Signal-to-noise ratio (SNR) was calculated and compared using semiparametrical multiple Conover-comparisons. Correlations between SNR and prostate volume and BMI were indicated using Pearson correlation coefficient. Distribution of SNR was evaluated for prostate quadrants. ResultsDiagnostic suitable spectra were achieved in 76 % (protocol 1, 100% in CSL), 32 % (protocol 2, 59% in CSL) and 50 % (protocol 3, 80% in CSL) of the voxels. SNR was significantly higher in protocol 3 compared to protocol 2 and 1 (93,729 vs. 27,836 vs. 32,897, p<0.0001) with significant difference between protocol 2 and 1 (p<0.023). Highest SNR was achieved in the dorsal prostate (protocols 1 and 3; p<0.0001). Sensitivity at 3.0T was higher with use of ERC. Specificity was highest at 1.5T with ERC. ConclusionThe ERC improves the diagnostic suitability and the SNR in MRSI at 3.0T. Less voxels at 3.0T with disabled ERC are suitable for diagnosis compared to 1.5T with ERC. MRSI at 3.0T with ERC shows the highest SNR. SNR in dorsal quadrants of the prostate was higher using ERC.

Metabolic imaging of pancreatic ductal adenocarcinoma detects altered choline metabolism.

Pancreatic ductal adenocarcinoma (PDAC) is an aggressive and lethal disease that develops relatively symptom-free and is therefore advanced at the time of diagnosis. The absence of early symptoms and effective treatments has created a critical need for identifying and developing new noninvasive biomarkers and therapeutic targets. We investigated the metabolism of a panel of PDAC cell lines in culture and noninvasively in vivo with (1)H magnetic resonance spectroscopic imaging (MRSI) to identify noninvasive biomarkers and uncover potential metabolic targets. We observed elevated choline-containing compounds in the PDAC cell lines and tumors. These elevated choline-containing compounds were easily detected by increased total choline (tCho) in vivo, in spectroscopic images obtained from tumors. Principal component analysis of the spectral data identified additional differences in metabolites between immortalized human pancreatic cells and neoplastic PDAC cells. Molecular characterization revealed overexpression of choline kinase (Chk)-α, choline transporter 1 (CHT1), and choline transporter-like protein 1 (CTL1) in the PDAC cell lines and tumors. Collectively, these data identify new metabolic characteristics of PDAC and reveal potential metabolic targets. Total choline detected with (1)H MRSI may provide an intrinsic, imaging probe-independent biomarker to complement existing techniques in detecting PDAC. The expression of Chk-α, CHT1, and CTL1 may provide additional molecular markers in aspirated cytological samples.

Abstract 1185: Altered choline metabolism in pancreatic cancer cells and tumor xenografts

Abstract Early-stage pancreatic cancer is usually clinically silent, and most patients presenting symptoms attributable to pancreatic cancer have advanced disease. Better understanding of the disease, effective detection methods, and new therapeutic targets are urgently needed to improve its outcome. Magnetic resonance spectroscopy (MRS) is being evaluated in the diagnosis of several solid malignancies (1). A hallmark of most solid tumors is the detection of elevated levels of phosphocholine (PC) and total choline (tCho) (1). The increase in PC levels in solid tumors has been attributed to high level of choline kinase (Chk)-alpha (2). Here, we characterized the metabolism of multiple pancreatic ductal adenocarcinoma cell lines and tumors with in vitro high resolution MRS and in vivo 1H magnetic resonance spectroscopic imaging (MRSI). We also characterized the expression of Chk-alpha in the cells and tumor extracts. Eight pancreatic adenocarcinoma cell lines (Panc1, BxPC3, Pa04C, Pa02C, Pa20C, Pa28C, Pa03C, and Pa09C) and one immortalized pancreatic cell line (human pancreatic nestin expressing (HPNE)) were investigated. For the in vivo experiments, cells were inoculated subcutaneously in male SCID mice. Once tumors reached 500 mm3, the mice were scanned on a 4.7T spectrometer for 1H MRSI, after which the tumors excised for immunoblot analysis, and high-resolution 1H MRS. Cells and tumor extracts were carried out as previously described (3). Fully relaxed 1H MR spectra of the extracts were acquired on a Bruker Avance 500 spectrometer at 11.7 T. Integrals of the metabolites of interest were determined and normalized to the number of cells and to the tumor weight respectively. Significant differences in tCho were identified in pancreatic cell lines and the tumor xenografts. In the panel of cell lines, the differences in tCho levels were associated with differences in Chk-alpha expression. PC and tCho levels were significantly elevated in the pancreatic cancer cell lines compared to the immortalized pancreatic cell line, and Chk-alpha was overexpressed in the cancer cell lines. In vivo studies revealed that elevated tCho was detected in tumors derived from the 4 cell lines tested, with higher levels of tCho in Panc1 tumors. These in vivo results were confirmed by ex vivo MRS analysis of the tumor extracts. The high level of tCho observed in Panc1 tumors was mainly due to increased PC, and correlated with high Chk-alpha expression observed in the immunoblots. These data support the use of 1H MRS to noninvasively detect pancreatic cancer. Moreover, the aberrant choline metabolism may provide novel targets in the treatment of pancreatic cancer. Acknowledgement: This work was supported by NIH P50CA103175. *Equal contribution.

Manifold Learning in MR spectroscopy using nonlinear dimensionality reduction and unsupervised clustering.

To investigate whether nonlinear dimensionality reduction improves unsupervised classification of (1) H MRS brain tumor data compared with a linear method. In vivo single-voxel (1) H magnetic resonance spectroscopy (55 patients) and (1) H magnetic resonance spectroscopy imaging (MRSI) (29 patients) data were acquired from histopathologically diagnosed gliomas. Data reduction using Laplacian eigenmaps (LE) or independent component analysis (ICA) was followed by k-means clustering or agglomerative hierarchical clustering (AHC) for unsupervised learning to assess tumor grade and for tissue type segmentation of MRSI data. An accuracy of 93% in classification of glioma grade II and grade IV, with 100% accuracy in distinguishing tumor and normal spectra, was obtained by LE with unsupervised clustering, but not with the combination of k-means and ICA. With (1) H MRSI data, LE provided a more linear distribution of data for cluster analysis and better cluster stability than ICA. LE combined with k-means or AHC provided 91% accuracy for classifying tumor grade and 100% accuracy for identifying normal tissue voxels. Color-coded visualization of normal brain, tumor core, and infiltration regions was achieved with LE combined with AHC. The LE method is promising for unsupervised clustering to separate brain and tumor tissue with automated color-coding for visualization of (1) H MRSI data after cluster analysis.

(1) H 2D MRSI tissue type analysis of gliomas.

To decompose 1H MR spectra of glioma patients into normal and abnormal tissue proportions for tumor classification and delineation. Anatomical imaging and 1H magnetic resonance spectroscopic imaging data have been acquired from 11 grade II and 13 grade IV glioma patients. LCModel was used to decompose the magnetic resonance spectroscopic imaging data into normal brain, grade II, and grade IV tissue proportions using a tissue type basis set. Simulations were conducted to evaluate the accuracy of the methodology. Results were visualized using colormaps and abnormality contours showing tumor grade and extent. Simulations suggest that infiltrative tumor proportions as low as 20% can be identified at the typical 1H magnetic resonance spectroscopy signal-to-noise found in vivo. Tumor grading according to the highest estimated tumor grade within a lesion gave a classification accuracy of 86% discriminating between grade II and grade IV glioma. Voxels with significant proportions of tumor type spectra were found beyond the margins of contrast enhancement for most grade IV cases consistent with infiltration whereas the abnormality contours show that some tumors are confined within the hyperintensities shown by both post contrast T1 weighted and T2 weighted imaging. LCModel can be used to decompose 1H MR spectra into proportions of normal and abnormal tissue to identify tumor extent, infiltration, and overall grade.

A 1H-magnetic resonance spectroscopy imaging study on frontal gyrus, anterior cingulate cortex and hippocampus of patients with generalized anxiety disorder

Objective To study the feature of brain functional in front gyrus, anterior cingulate cortex and hippocampus of patients with generalized anxiety disorder (GAD). Methods 19 patients with GAD and 20 healthy volunteers were scanned on brain using proton magnetic resonance spectroscopic imaging (1H-MRS). The levels of Choline (Cho), Creatine (Cr), N-acetyl-aspartate(NAA) were measured in the frontal gyrus, anterior cingulate cortex and hippocampus of all subjects.The 1H-MRS data were compared between two groups. Results Compared with the healthy matched control, the levels of Cho(7.22±1.99), Cr(5.44±1.68), NAA(12.09±2.30)in right frontal gyrus white matter, the levels of Cho(9.89±2.40), Cr(8.59±1.71)in right anterior cingulate cortex and the levels of NAA in left anterior cingulate cortex were significantly high (P 0.05). Conclusion The brain substance metabolisms of the patients with GAD are abnormal and asymmetrical between left and right brain, especially occurred in right brain. Key words: 1H-Magnetic resonance spectroscopy; Generalized anxiety disorder; Frontal gyrus; Anterior cingulate cortex; Hippocampus

Neurochemical abnormalities in unmedicated bipolar depression and mania: A 2D 1H MRS investigation

The neurobiology and neurochemistry of bipolar disorder and its different phases are poorly understood. This study investigated metabolite abnormalities in both unmedicated bipolar depression as well as mania using 2D 1H magnetic resonance spectroscopy imaging (MRSI). MRSI data were obtained from 24 unmedicated bipolar disorder (BP) subjects (12 (hypo)manic (BPM)) and 12 depressed (BPD), and 20 closely matched healthy controls. 2D 1H MRSI data were collected from a 15-mm axial slice placed along the anterior commissure-posterior commissure (AC-PC) line to measure brain metabolites bilaterally in the thalamus and also the anterior and posterior cingulate cortex (ACC and PCC). Brain Lac/Cr levels were significantly increased in the BP group as a whole compared to healthy controls. Glutamate abnormalities varied across bipolar state as well as brain region: significantly increased Glx/Cr values were found in the left thalamus in BPD, but BPM had decreased Glu/Cr and Glx/Cr levels in the PCC when compared to healthy controls and decreased Glu/Cr levels even when compared to the BPD subjects group. The findings of the study point to state-related abnormalities of oxidative and glutamate metabolism in bipolar disorder.

WE-C-116-12: Camer-Rao Lower Bound Analysis of 1H Magnetic Resonance Spectroscopic Imaging for Breast with Singular Value Decomposition, LCModel and QUEST

Purpose: LCModel and QUEST supply error estimate capability, however no standard basis available for unknown lipids mixture in breast. Singular Value Decomposition (SVD) with Linear Combination method was developed to implement desired CRLB function. The purpose of this study is to evaluate SVD, QUEST and LCModel quantification for breast MRSI with Camer-Rao Lower Bound (CRLB) analysis. Methods: Spectrum for 6 metabolite were simulated with Vespa (STEAM SVS, 3.0T, TE=42ms, TM=10ms), mixture spectrum was simulate with known ratio. jMRUI 4.0 (QUEST function) and LCModel 6.3 was used in the study, metabolite basis was generated from individually simulated data. In-house SVD was developed with Matlab 7.6 to combine Hankel SVD and Linear Combination method. The estimated FID signals are constructed from the quantification parameters, residue error spectrum was calculated by FFT of the difference of original FID and estimated FID. In order to study the CRLB, known simulated spectrum without noise was analyzed with SVD, QUEST and LCModel. Then introduce random noise, quantification residue was analyzed, and estimated parameters with CRLB were compared with results without noise. Sample patient breast MRSI data was collected with Siemens TrioTim (PRESS CSI, 3.0T, TE=80ms). Results: For simulated data, both Quest and LCModel can reliably quantify with SNR of 5 and signal as low as 17% of maximum; while SVD method can quantify with SNR of 10.Representative breast MRSI data were analyzed, LCModel residue error is 13.3% of original maximum and Cho CRLB is 56%; while SVD can achieve 3.7% and 3.6% respectively. Conclusions: For all methods, CRLB estimations increase with decrease the SNR. SVD is sensitive to noise, since it has no prior knowledge to reduce noise, if noise is larger than signal, fake peak can be introduced; while LCModel and QUEST can not quantify breast data well with limited simulated lipid basis.

MR Spectroscopic Imaging of Peripheral Zone in Prostate Cancer Using a 3T MRI Scanner: Endorectal versus External Phased Array Coils.

Magnetic resonance spectroscopic imaging (MRSI) detects alterations in major prostate metabolites, such as citrate (Cit), creatine (Cr), and choline (Ch). We evaluated the sensitivity and accuracy of three-dimensional MRSI of prostate using an endorectal compared to an external phased array “receive” coil on a 3T MRI scanner. Eighteen patients with prostate cancer (PCa) who underwent endorectal MR imaging and proton (1H) MRSI were included in this study. Immediately after the endorectal MRSI scan, the PCa patients were scanned with the external phased array coil. The endorectal coil-detected metabolite ratio [(Ch+Cr)/Cit] was significantly higher in cancer locations (1.667 ± 0.663) compared to non-cancer locations (0.978 ± 0.420) (P < 0.001). Similarly, for the external phased array, the ratio was significantly higher in cancer locations (1.070 ± 0.525) compared to non-cancer locations (0.521 ± 0.310) (P < 0.001). The sensitivity and accuracy of cancer detection were 81% and 78% using the endorectal ‘receive’ coil, and 69% and 75%, respectively using the external phased array ‘receive’ coil.

1H MRSI of middle frontal gyrus in pediatric ADHD

Neuroimaging studies in multiple modalities have implicated the left or right dorsolateral prefrontal cortex (here, middle frontal gyrus) in attentional functions, in ADHD, and in dopamine agonist treatment of ADHD. The far lateral location of this cortex in the brain, however, has made it difficult to study with magnetic resonance spectroscopy (MRS). We used the smaller voxel sizes of the magnetic resonance spectroscopic imaging (MRSI) variant of MRS, acquired at a steep coronal-oblique angle to sample bilateral middle frontal gyrus in 13 children and adolescents with ADHD and 13 age- and sex-matched healthy controls. Within a subsample of the ADHD patients, aspects of attention were also assessed with the Trail Making Task. In right middle frontal gyrus only, mean levels of N-acetyl-aspartate + N-acetyl-aspartyl-glutamate (tNAA), creatine + phosphocreatine (Cr), choline-compounds (Cho), and myo-inositol (mI) were significantly lower in the ADHD than in the control sample. In the ADHD patients, lower right middle frontal Cr was associated with worse performance on Trails A and B (focused attention, concentration, set-shifting), while the opposite relationship held true for the control group on Trails B. These findings add to evidence implicating right middle frontal cortex in ADHD. Lower levels of these multiple species may reflect osmotic adjustment to elevated prefrontal cortical perfusion in ADHD and/or a previously hypothesized defect in astrocytic production of lactate in ADHD resulting in decelerated energetic metabolism (Cr), membrane synthesis (Cho, mI), and acetyl-CoA substrate for NAA synthesis. Lower Cr levels may indicate attentional or executive impairments.

TU‐G‐217A‐08: Camer‐Rao Lower Bound Analysis of 1H Magnetic Resonance Spectroscopic Imaging for Breast With Singular Value Decomposition Method and Linear Combination Method

Purpose: To study the Singular Value Decomposition (SVD) and Linear Combination (LC) quantification with Camer‐Rao Lower Bound (CRLB) analysis. Methods: In‐house Singular Value Decomposition (SVD) and LC program was developed with Matlab version 7.6 (Mathworks.com). Hankel SVD method (singular value decomposition of the acquired FID signal arranged in a Hankel matrix) is used to compute the signal poles and amplitude, and from them the signal frequencies and damping factors. In LC method, poles are generated from SVD analysis of averaged 10 FID, the amplitudes of the signal components were estimated by the least square method for individual FID signal. The estimated FID signals are constructed from the quantification parameters, residue error spectrum was calculated by FFT of the difference of original FID and estimated FID. In order to study the CR lower bound, known simulated spectrum without noise was analyzed with SVD and LC method. Then introduce random noise, quantification residue was analyzed, and estimated parameters with CRLB were compared with results without noise. Results: CRLB estimation is verified with the standard deviation calculation, CRLB is larger than and comparable to standard deviation. For small signal as low as 1.5% of maximum, SVD can detect at SNR 10, LC can handle SNR 5. For SNR 5, SVD can reliably detect signal like 12.5% of maximum. Typical patient data from breast tumor data was analyzed with SVD and LC methods too, results are reliable and consistent. Conclusions: CRLB estimation is verified with the standard deviation calculation. Both SVD and LC method can be used for human MRSI, the SVD method is sensitive to noise, since it has no prior knowledge to reduce noise; if noise is larger than signal, fake peak can be introduced. LC method combined with SVD generated poles can be exploited in MRSI with best results.

Automated vs. Manual Pattern Recognition of 3D 1H MRSI Data of Patients with Prostate Cancer

The aim of this study was to assess (1) automated analysis methods versus manual evaluation by human experts of three-dimensional proton magnetic resonance spectroscopic imaging (MRSI) data from patients with prostate cancer and (2) the contribution of spatial information to decision making. Three-dimensional proton MRSI was applied at 1.5 T. MRSI data from 10 patients with histologically proven prostate adenocarcinoma, scheduled either for prostatectomy or intensity-modulated radiation therapy, were evaluated. First, two readers manually labeled spectra using spatial information to identify the localization of spectra and neighborhood information, establishing the reference set of this study. Then, spectra were labeled again manually in a blinded and randomized manner and evaluated automatically using software that applied spectral line fitting as well as pattern recognition routines. Statistical analysis of the results of the different approaches was performed. Altogether, 1018 spectra were evaluable by all methods. Numbers of evaluable spectra differed significantly depending on patient and evaluation method. Compared to automated analysis, the readers made rather binary decisions, using information from neighboring spectra in ambiguous cases, when evaluating MRSI data as a whole. Differences between anatomically blinded and unblinded evaluation were larger than differences between evaluations using blinded data and automated techniques. An automated approach, which evaluates each spectrum individually, can be as good as an anatomy-blinded human reader. Spatial information is routinely used by human experts to support their final decisions. Automated procedures that consider anatomic information for spectral evaluation will enhance the diagnostic impact of MRSI of the human prostate.

Skeletal muscle 1H MRSI before and after prolonged exercise. I. muscle specific depletion of intramyocellular lipids

Aim of the study was to determine distribution and depletion patterns of intramyocellular lipids (IMCL) in leg muscles before and after two types of standardized endurance exercise. ¹H-magnetic resonance spectroscopic imaging was performed (1) in the thigh of eight-trained cyclists after exercising on an ergometer for 3 h at 52 ± 8% of maximal speed and (2) in the lower leg of eight-trained runners after exercising on a treadmill for 3 h at 49 ± 3% of maximal workload. Pre-exercise IMCL contents were reduced postexercise in 11 out of 13 investigated upper and lower leg muscles (P < 0.015 for all). A strong linear correlation with a slope of ∼0.5 between pre-exercise IMCL content and IMCL depletion was found. IMCL depletion differed strongly between muscles. Absolute and also relative IMCL reduction was significantly higher in muscles with predominantly slow fibers compared to those with fast fibers. Creatine levels and fiber orientation were stable and unchanged after exercise, while trimethyl-ammonium groups increased. This is presented in the accompanying paper. In conclusion, a systematic comparison of metabolic changes in cross sections of the upper and lower leg was performed. The results imply that pre-exercise IMCL levels determine the degree of IMCL depletion after exercise.

1H MRSI and social communication deficits in pediatric complex partial seizures

To investigate relationships between regional brain metabolites, social communication deficits, and seizure frequency in children and adolescents with cryptogenic epilepsy with complex partial seizures (CPS). In 12 children and adolescents with CPS and 23 age- and gender-matched healthy controls, we acquired proton magnetic resonance spectroscopic imaging (MRSI) at 1.5 T and 30 ms echo-time from bilateral inferior frontal and superior temporal gyri, regions associated with social communication deficits. Videotaped speech samples of all the subjects were coded for social communication deficits and parents provided information on seizure frequency. Four MRSI findings emerged in right inferior frontal gyrus. N-acetyl-aspartate (NAA) plus N-acetyl-aspartyl-glutamate (NAAG)--together called "tNAA"--was 11.4% lower in patients with CPS than in controls. Choline-compounds (Cho) were 15.4% lower in CPS than in controls. Within CPS, higher tNAA was associated with more frequent seizures and abnormal social communication. Localization of findings to right inferior frontal cortex supports the involvement of this area in social communication deficits and may be related to atypical lateralization of expressive language in pediatric epilepsy. Lower levels of tNAA and Cho may indicate local neuronal or glial damage or underpopulation due to excitotoxicity or other causes. The sensitivity of tNAA to seizure frequency suggests effects of ongoing CPS on neuronal and glial function in this brain region.

SU‐E‐I‐123: Quantification of 1H Magnetic Resonance Spectroscopic Imaging for Breast Cancer with Singular Value Decompsition (SVD) Method

Purpose: To study the feasibility of Singular Value Decomposition (SVD) method to quantify the MRSI data, and the effect of signal to noise ratio on SVD method was studied. Methods: In‐house SVD program was developed with Matlab version 7.6 (Mathworks.com), in frequency domain water signal peak around 4.7ppm is used for referencing shifting. Hankel SVD method is used to compute the signal poles and from them the signal frequencies and damping factors. Known spectrum without noise was analyzed. Then introduce random noise, Gaussian filter of various frequency was used to reduce noise effect, wavelet threshold de‐noise method was also tested with various hard and soft threshold selection methods. Human breast MRSI data were analyzed with optimized parameters.Results: SVD can easily detect the small signal in low noise GSH data at 2.54ppm, GSH signal ratio to NAA is only 0.5%. However SVD can not detect the signal after added random noise of SNR 15. By add Gaussian filter of 2.5Hz, peak could be detected by suppress noise, however if noise level is larger than the signal, fake peak could be resulted. For wavelet threshold method, hard threshold with wavelet principle of Steinˈs Unbiased Risk Estimate (SURE) can generate best results, random noise of SNR 7 can be reliably detected. Typical patient data from breast tumor data was analyzed, even the detected cho peak is only 0.5% of water peak, it can be detected reliably, and the max residue error is 0.2% of water peak. Conclusions: SVD method can be used for human breast MRSI. the SVD method is sensitive to detect the signals as low as 0.5%. However it is suspected to noise, since it has no prior knowledge to reduce noise; improved SVD method with wavelet threshold de‐noise can exploited in MRSI with SNR of 7.