Berthold Kiefer

Generalized autocalibrating partially parallel acquisitions (GRAPPA)

In this study, a novel partially parallel acquisition (PPA) method is presented which can be used to accelerate image acquisition using an RF coil array for spatial encoding. This technique, GeneRalized Autocalibrating Partially Parallel Acquisitions (GRAPPA) is an extension of both the PILS and VD‐AUTO‐SMASH reconstruction techniques. As in those previous methods, a detailed, highly accurate RF field map is not needed prior to reconstruction in GRAPPA. This information is obtained from several k‐space lines which are acquired in addition to the normal image acquisition. As in PILS, the GRAPPA reconstruction algorithm provides unaliased images from each component coil prior to image combination. This results in even higher SNR and better image quality since the steps of image reconstruction and image combination are performed in separate steps. After introducing the GRAPPA technique, primary focus is given to issues related to the practical implementation of GRAPPA, including the reconstruction algorithm as well as analysis of SNR in the resulting images. Finally, in vivo GRAPPA images are shown which demonstrate the utility of the technique. Magn Reson Med 47:1202–1210, 2002.

The Virtual Family—development of surface-based anatomical models of two adults and two children for dosimetric simulations

The objective of this study was to develop anatomically correct whole body human models of an adult male (34 years old), an adult female (26 years old) and two children (an 11-year-old girl and a six-year-old boy) for the optimized evaluation of electromagnetic exposure. These four models are referred to as the Virtual Family. They are based on high resolution magnetic resonance (MR) images of healthy volunteers. More than 80 different tissue types were distinguished during the segmentation. To improve the accuracy and the effectiveness of the segmentation, a novel semi-automated tool was used to analyze and segment the data. All tissues and organs were reconstructed as three-dimensional (3D) unstructured triangulated surface objects, yielding high precision images of individual features of the body. This greatly enhances the meshing flexibility and the accuracy with respect to thin tissue layers and small organs in comparison with the traditional voxel-based representation of anatomical models. Conformal computational techniques were also applied. The techniques and tools developed in this study can be used to more effectively develop future models and further improve the accuracy of the models for various applications. For research purposes, the four models are provided for free to the scientific community.

Double echo steady state magnetic resonance imaging of knee articular cartilage at 3 Tesla: a pilot study for the Osteoarthritis Initiative.

Background: Quantitative magnetic resonance imaging (qMRI) may provide valuable measures of cartilage morphology in osteoarthritis (OA) but has been confined to sequences with relatively long acquisition times at 1.5 Tesla (T). Objective: To test the accuracy and precision of knee cartilage qMRI with a fast double echo, steady state (DESS) sequence with water excitation (we) at 3 T. Methods: As a pilot study for the Osteoarthritis Initiative, test-retest MR images were acquired in the knees of 19 participants with no OA to moderate degrees of clinical OA. Two double oblique coronal fast low angle shot (FLASHwe) sequences (1.5 mm slice thickness) were acquired at 3 T, and two sagittal DESSwe sequences (0.7 mm slice thickness). Double oblique coronal multiplanar reformats (MPR) were performed (1.5 mm slice thickness) from the sagittal DESSwe. Knee joint cartilage plates were quantified unpaired in random order with blinding to subject identification. Results: In the femorotibial joint, precision errors (root mean square coefficient of variation in % for unpaired analysis) for cartilage volume and thickness were 3.0–6.4% with coronal FLASHwe, 2.4–6.2% with coronal MPR DESSwe, and 2.3–8.2% with sagittal DESSwe. Correlation coefficients between DESSwe and FLASHwe ranged from r = 0.88 to 1.0. In the femoropatellar joint, precision errors (sagittal DESSwe) were 3.4–8.5%. Conclusions: DESSwe permits accurate and precise analysis of cartilage morphology in the femorotibial joint at 3 T. Further studies are needed to examine the accuracy of DESSwe in the femoropatellar joint and its ability to characterise sensitivity to longitudinal changes in cartilage morphology.

Magnetic resonance imaging of the body trunk using a single-slab, 3-dimensional, T2-weighted turbo-spin-echo sequence with high sampling efficiency (SPACE) for high spatial resolution imaging: Initial clinical experiences

Purpose:The authors conducted a clinical evaluation of single-slab, 3-dimensional, T2-weighted turbo-spin-echo (TSE) with high sampling efficiency (SPACE) for high isotropic body imaging with large field-of-view (FoV). Materials and Methods:Fifty patients were examined in clinical routine with SPACE (regions of interest: pelvis n = 30, lower spine n = 12, upper spine n = 6, extremities n = 4) at 1.5 T. For achieving a high sampling efficiency, parallel imaging, high turbofactor, and magnetization restore pulses were used. In contrast to a conventional TSE imaging technique with constant flip angle refocusing, the refocusing pulse train of the SPACE sequence consists of variable flip angle radiofrequency pulses along the echo train. Results:Signal-to-noise ratio and contrast-to-noise ratio of SPACE images were of sufficient diagnostic value. The possibility of image reconstruction in multiple planes was of clinical relevance in all cases and simplified data analysis. Conclusion:The achievement of 3-dimensional, T2-weighted TSE magnetic resonance imaging with isotropic and high spatial resolution and interactive 3-dimensional visualization essentially improve the diagnostic potential of magnetic resonance imaging.

Free-breathing radial 3D fat-suppressed T1-weighted gradient echo sequence: A viable alternative for contrast-enhanced liver imaging in patients unable to suspend respiration

Objective:To compare free-breathing radially sampled 3D fat suppressed T1-weighted gradient-echo acquisitions (radial volumetric interpolated breath-hold examination [VIBE]) with breath-hold (BH) and free-breathing conventional (rectilinearly sampled k-space) VIBE acquisitions for postcontrast imaging of the liver. Materials and Methods:Eighteen consecutive patients referred for clinically indicated liver magnetic resonance imaging were imaged at 3 T. Three minutes after a single dose of gadolinium contrast injection, free-breathing radial VIBE, BH VIBE, and free-breathing VIBE with 4 averages were acquired in random order with matching sequence parameters. Radial VIBE was acquired with the “stack-of-stars” scheme, which uses conventional sampling in the slice direction and radial sampling in-plane. All image data sets were evaluated independently by 3 radiologists blinded to patient and sequence information. Each reader scored the following parameters: overall image quality, respiratory motion artifact, pulsation artifact, liver edge sharpness, and hepatic vessel clarity using a 5-point scale, with the highest score indicating the most optimum examination. Mixed model analysis of variance was used to compare sequences in terms of each measure of image quality. Results:When scores were averaged over readers, there was no statistically significant difference between radial VIBE and BH VIBE regarding overall image quality (P = 0.1015), respiratory motion artifact (P = 1.0), and liver edge sharpness (P = 0.2955). Radial VIBE demonstrated significantly lower pulsation artifact (P < 0.0001), but had lower hepatic vessel clarity (P = 0.0176), when compared with BH VIBE. Radial VIBE had significantly higher image quality scores for all parameters when compared with free-breathing VIBE (P < 0.0001). Acquisition time for BH VIBE was 14 seconds and that of free-breathing radial VIBE and conventional VIBE with multiple averages was 56 seconds each. Conclusion:Radial VIBE can be performed during free breathing for contrast-enhanced imaging of the liver with comparable image quality to BH VIBE. However, further work is necessary to shorten the acquisition time to perform dynamic imaging.

Self-gated Radial MRI for Respiratory Motion Compensation on Hybrid PET/MR Systems

Accurate localization and uptake quantification of lesions in the chest and abdomen using PET imaging is challenging due to the respiratory motion during the exam. The advent of hybrid PET/MR systems offers new ways to compensate for respiratory motion without exposing the patient to additional radiation. The use of self-gated reconstructions of a 3D radial stack-of-stars GRE acquisition is proposed to derive a high-resolution MRI motion model. The self-gating signal is used to perform respiratory binning of the simultaneously acquired PET raw data. Matching mu-maps are generated for every bin, and post-reconstruction registration is performed in order to obtain a motion-compensated PET volume from the individual gates. The proposed method is demonstrated in-vivo for three clinical patients. Motion-corrected reconstructions are compared against ungated and gated PET reconstructions. In all cases, motion-induced blurring of lesions in the liver and lung was substantially reduced, without compromising SNR as it is the case for gated reconstructions.

Diagnosis of Hepatic Metastasis: Comparison of Respiration-Triggered Diffusion-Weighted Echo-Planar MRI and Five T2-Weighted Turbo Spin-Echo Sequences

OBJECTIVE The purpose of this study was to compare the value of respiration-triggered diffusion-weighted (DW) single-shot echo-planar MRI (EPI) and five variants of T2-weighted turbo spin-echo (TSE) sequences in the diagnosis of hepatic metastasis. MATERIALS AND METHODS Fifty-two patients with extrahepatic primary malignant tumors underwent 1.5-T MRI that included DW EPI and the following variants of T2-weighted TSE techniques: breath-hold fat-suppressed HASTE, breath-hold fat-supressed TSE, respiration-triggered fat-suppressed TSE, breath-hold STIR, and respiration-triggered STIR. Images were reviewed independently by two blinded observers who used a 5-point confidence scale to identify lesions. Results were correlated with surgical and histopathologic findings and follow-up imaging findings. The accuracy of each technique was measured with free-response receiver operating characteristic analysis. RESULTS A total of 118 hepatic metastatic lesions (mean diameter, 12.8 mm; range, 3-84 mm) were evaluated. Accuracy values were higher (p < 0.001) with DW EPI (0.91-0.92) than with the T2-weighted TSE techniques (0.47-0.67). Imaging with the HASTE sequence (0.47-0.52) was less accurate (p < 0.05) than imaging with the breath-hold TSE, breath-hold STIR, respiration-triggered TSE, and respiration-triggered STIR sequences (0.59-0.67). Sensitivity was higher (p < 0.001) with DW EPI (0.88-0.91) than with T2-weighted TSE techniques (0.45-0.62). For small (< or = 10 mm) metastatic lesions only, the differences in sensitivity between DW EPI (0.85) and T2-weighted TSE techniques (0.26-0.44) were even more pronounced. CONCLUSION DW EPI was more sensitive and more accurate than imaging with T2-weighted TSE techniques. Because of the black-blood effect on vessels and low susceptibility to motion artifacts, DW EPI was particularly useful for the detection of small (< or = 10 mm) metastatic lesions.

Assessment of normal patellar cartilage volume and thickness using MRI : an analysis of currently available pulse sequences

Abstract Objective. The objective of this study was to analyse the potential of magnetic resonance imaging for valid determination of patellar cartilage thickness, comparing currently available pulse sequences. Design. In six patients and one cadaver the cartilage was repetitively imaged employing three spin-echo and six three-dimensional gradient-echo sequences. In the cadaveric specimen the total volume and the regional distribution of cartilage thickness were assessed and compared with the values obtained from anatomical sections by image analysis. Results and conclusions. The FLASH and fat-suppressed FLASH sequences allowed the most accurate determination of the cartilage volume and thickness. Fat-suppression considerably increased the contrast of the cartilage to the synovial fluid, fat and bone marrow, yielding higher reproducibility of the volumetric measurements. The remaining difference from the anatomical volume and thickness may be because the calcified cartilage is not delineated by magnetic resonance imaging.

Tumor detection by Diffusion Weighted MRI and ADC-Mapping - Initial Clinical Experiences in Comparison to PET-CT

Objective:To evaluate the clinical potential of diffusion-weighted-imaging (DWI) with apparent diffusion coefficient (ADC)-mapping for tumor detection. Materials and Methods:A single-shot echo-planar-imaging DWI sequence with fat suppression and ability for navigator-based respiratory triggering was implemented. Nineteen patients (11 melanoma, 4 prostate cancer, 1 non-Hodgkin lymphoma, and 3 lung cancer) were examined by positron emission tomography (PET) with an integrated computed tomography scanner (PET-CT) and DWI. Images at b = 0, 400, and 1000 s/mm2 were acquired and ADC maps were generated. PET examinations were used as a reference for tumor detection. Four hundred twenty-four regions of interest were used for DWI and 73 for PET data evaluation. Results:DWI and ADC maps were of diagnostic quality. Metastases with increased tracer uptake were clearly visualized at b = 1000 s/mm2 with the exception of mediastinal lymph node metastases in cases of lung cancer. ADC mapping did not improve detection rates. Conclusions:DWI is a feasible clinical technique, improving the assessment of metastatic spread in routine magnetic resonance imaging examinations.

Intravoxel Incoherent Motion Diffusion-weighted MR Imaging for Characterization of Focal Pancreatic Lesions

PURPOSE To evaluate the diagnostic potential of apparent diffusion coefficient (ADC) and intravoxel incoherent motion (IVIM)-derived parameters for differentiation of common pancreatic tumors, chronic pancreatitis, and normal pancreas and for characterization of the malignancy potential of intraductal papillary mucinous neoplasms (IPMNs). MATERIALS AND METHODS The institutional review board approved this retrospective study, and informed consent was waived. Ninety-three consecutive patients with surgically resected and pathologically confirmed pancreatic tumors (39 pancreatic adenocarcinomas [PACs], 17 neuroendocrine tumors [NETs], and 37 IPMNs), seven patients with chronic pancreatitis, and 26 patients with a normal pancreas were included in this study. All patients underwent pancreatic 3.0-T magnetic resonance imaging, including IVIM diffusion-weighted imaging with 10 b values used (from 0 to 1000 sec/mm(2)). The ADC, slow component of diffusion (D(slow)), incoherent microcirculation (D(fast)), and perfusion fraction (f) were calculated. Steel-Dwass and Mann-Whitney U tests were used for comparison. The diagnostic performance of the parameters was evaluated by using receiver operating characteristic (ROC) analysis with Bonferroni correction. RESULTS Among ADC- and IVIM-derived parameters, D(fast) and f values of PACs were significantly lower than those of normal pancreas, chronic pancreatitis, and NETs (all P < .05 in post hoc analyses). For differentiation of PACs from NETs, f and D(fast) showed a significant difference (P < .0001 for both) and were more useful parameters than ADC and D(slow) in ROC analysis (all P < .05). Malignant IPMNs had significantly lower ADC and D(slow) values and higher D(fast) and f values when compared with benign IPMNs (all P < .05). In ROC analysis, f showed the highest area under the ROC curve value for distinguishing malignant from benign IPMNs. CONCLUSION IVIM-derived perfusion-related parameters could be helpful for the differentiation of common malignant tumors in the pancreas and for distinguishing malignant from benign IPMNs. D(fast) and f were more valuable parameters in the differentiation of PACs from NETs than were ADC and D(slow).