Electronic poster

Wednesday 13:30-15:30 Computer 100

13:30 4702. Whole-Body MRI, Including Diffusion-Weighted Imaging, for Staging Malignant Lymphoma: Direct Comparison to CT

Thomas Kwee¹, Malou Vermoolen¹, Erik Akkerman², Henriëtte Quarles van Ufford¹, Frederik Beek¹, Inge Ludwig³, Marc Bierings⁴, Rob Fijnheer⁵, Marie-José Kersten⁶, Joseph Zsiros⁷, Willem Mali¹, Rutger-Jan Nievelstein<sup>1

1</sup>Department of Radiology, University Medical Center Utrecht, Utrecht, Netherlands; ²Department of Radiology, Academic Medical Center, Amsterdam, Netherlands; ³Department of Hematology, University Medical Center Utrecht, Utrecht, Netherlands; ⁴Department of Pediatric Hematology, University Medical Center Utrecht, Utrecht, Netherlands; ⁵Department of Hematology, Meander Medical Center, Amersfoort, Netherlands; ⁶Department of Hematology, Academic Medical Center, Amsterdam, Netherlands; ⁷Department of Pediatric Hematology, Academic Medical Center, Amsterdam, Netherlands

This study aimed to assess the equivalence of whole-body MRI, including diffusion-weighted imaging (DWI), to computed tomography (CT) for the initial staging of malignant lymphoma. To that end, 66 consecutive patients with newly diagnosed malignant lymphoma prospectively underwent whole-body MRI (T1-weighted and short inversion time inversion recovery [n=66], and DWI [n=62]) at 1.5T and CT. Whole-body MRI (both with and without DWI) was equal to staging using CT in the majority of patients, while whole-body MRI overstaging occurs more frequently than whole-body MRI understaging (relative to CT).

14:00 4703. Contrast Edited T₂ SSFSE: A Potential Alternative to Diffusion Weighted EPI for MRI Screening

Ananth J. Madhuranthakam¹, Aya Yassin^2,3, Jean H. Brittain⁴, David C. Alsop^2,3, Neil M. Rofsky<sup>2,3

1</sup>MR Applied Science Lab, GE Healthcare, Boston, MA, United States; ²Radiology, Beth Israel Deaconess Medical Center, Boston, MA, United States; ³Harvard Medical School, Boston, MA, United States; ⁴MR Applied Science Lab, GE Healthcare, Madison, WI, United States

MRI has gained increased attention for whole-body screening of tumor metastasis. The commonly used sequences are STIR and DW-EPI. However, both are SNR limited and require multiple signal averages increasing the total scan time. In addition, DW-EPI images are also subject to distortion in larger FOVs, and are typically limited to low resolution, axial plane imaging. The primary objective of whole-body screening is to minimize the background tissue signal while simultaneously highlighting tumor signal in a rapid acquisition. Such a technique is presented in this work using SSFSE readout while suppressing signal from fat, fluid and blood vessels.

14:30 4704. Assessing the Advantages of Whole Body Diffusion Tensor Imaging for Screening of Bone Metastases

Matthew David Blackledge¹, David J. Collins¹, Toni Wallace¹, Dow-Mu Koh¹, Martin O. Leach<sup>1

1</sup>CR-UK and EPSRC Cancer Imaging Centre, Institute of Cancer Research and Royal Marsden Hospital, Sutton, Surrey, United Kingdom

We have demonstrated that the use of whole body DTI is feasible and practicable in the time required to obtain standard diffusion weighted whole body measurements of bone metastases without significant loss in image quality. Furthermore it is possible to measure useful biological parameters such as fractional anisotropy which may have direct clinical relevance in detecting spinal cord compression.

15:00 4705. ADC Measurements in the Evaluation of Lymph Nodes in Patients with Non-Hodgkin Lymphoma

Thomas Kwee¹, Inge Ludwig², Cuno Uiterwaal³, Henriëtte Quarles van Ufford¹, Marc Bierings⁴, Rob Fijnheer⁵, Taro Takahara¹, Rutger-Jan Nievelstein<sup>1

1</sup>Department of Radiology, University Medical Center Utrecht, Utrecht, Netherlands; ²Department of Hematology, University Medical Center Utrecht, Utrecht, Netherlands; ³Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht, Netherlands; ⁴Department of Pediatric Hematology, University Medical Center Utrecht, Utrecht, Netherlands; ⁵Department of Hematology, Meander Medical Center, Amersfoort, Netherlands

This study aimed to investigate whether apparent diffusion coefficient (ADC) measurements allow discriminating normal lymph nodes from lymphomatous lymph nodes, and indolent lymphomas from aggressive lymphomas in patients with non-Hodgkin lymphoma (NHL). To that end, 18 healthy volunteers and 22 patients with newly diagnosed NHL (indolent: n=9; aggressive: n=13) prospectively underwent diffusion-weighted imaging at 1.5 T. Our results suggest that ADC measurements may be a highly specific method for discriminating normal lymph nodes from lymphomatous lymph nodes in patients with NHL. However, ADC measurements appear to be of no utility in differentiating indolent from aggressive lymphomas.

Thursday 13:30-15:30 Computer 100

13:30 4706. Optimization of the B-Sampling for Bi-Exponential Analysis of Diffusion-Weighted Imaging

Jeff Lei Zhang¹, Eric E. Sigmund¹, Hersh Chandarana¹, Henry Rusinek¹, Hua Guo¹, Pippa Storey¹, Qun Chen¹, Vivian S. Lee<sup>1

1</sup>Department of Radiology, New York University, New York, NY, United States

Synopsis: To improve precision of bi-exponential parameters of diffusion weighted imaging (DWI), we proposed a method for optimizing b values for DWI acquisition. Monte Carlo simulation was performed to explore whether the optimized b values would improve the bi-exponential parameters’ ability in differentiating benign and malignant renal lesions. Results showed that, using the optimized b values, the differentiability of perfusion fraction for the two types of lesions improved by 22%±7%, compared with uniformly distributed b values. The method should be applicable to DWI of any other tissues or organs where bi-exponential analysis is used.

14:00 4707. Signal to Noise Ratio of High B-Value Diffusion Weighted Images Is Improved Using Computed Diffusion Weighted Imaging

Matthew David Blackledge¹, David J. Collins¹, Dow-Mu Koh¹, Martin O. Leach<sup>1

1</sup>CR-UK and EPSRC Cancer Imaging Centre, Institute of Cancer Research and Royal Marsden Hospital, Sutton, Surrey, United Kingdom

AAn acquisition scheme optimized for computed Diffusion Weighted Imaging (cDWI) is discussed and a theoretical model for diffusion weighted image noise is presented. It is demonstrated through theory and experimental studies that noise is reduced in calculated high b-value images compared to conventional acquired high b-value DWI using this method.

14:30 4708. Comparison of Liver ADC Measurements Using Breath-Hold, Free Breath-Hold and Respiratory Gating Echoplanar Diffusion-Weighted Imaging Sequences Using Parallel Imaging Technique with Different Acceleration Factors

Chun-Jung Juan¹, Hing-Chiu Chang^2,3, Hsiao-Wen Chung^1,3, Chi-Hong Chu⁴, Cheng-Chieh Cheng^1,3, Su-Chin Chiu^1,3, Hui-Chu Chiu^5,6, Cheng-Hsien Hsu⁷, Cheng-Yu Chen¹, Guo-Shu Huang<sup>1

1</sup>Department of Radiology, Tri-Service General Hospital, Taipei, Taiwan; ²Applied Science Laboratory, GE Healthcare Taiwan, Taipei, Taiwan; ³Institute of Biomedical Electronics and Bioinformatics, National Taiwan University, Taipei, Taiwan; ⁴Section of General Surgery, Department of Surgery, Tri-Service General Hospital, Taipei, Taiwan; ⁵Department of Nuclear Medicine, Tri-Service General Hospital, Taipei, Taiwan; ⁶EMBA in Global Chinese Management, Department of Business Administration, College of Management, Tamkang University, Taiwan; ⁷Division of Software Design, Notebook Unit 5, Quanta Computer Inc., Taiwan

Inter-experimental comparison of the liver apparent diffusion coefficient (ADC) value remains challenging. In this study the effect of repetition time (TR) and acceleration factors on the liver ADC measurement is examined. Our study shows that free-breath method allows ADC measurement consistent with breath-hold method with the TR. The ADC values measured at a longer TR in both free-breath and respiration-triggered methods are significantly higher than breath-hold method. Our results also depict significant differences in liver ADC values when different acceleration factors are chosen. Our results highlight the important of TR and acceleration factors in liver ADC measurements for inter-experimental comparison.

15:00 4709. Assessment of Variability of Region of Interest (ROI) Delineation on Diffusion Weighted MRI (DW-MRI) Using Manual and Semi-Automated Computer Methods

Nina Tunariu¹, James A. d'Arcy, Veronica A. Morgan¹, Michael Germuska, Catherine G. Simpkin, Sharon L. Giles², David J. Collins, Nandita M. deSouza

¹ CR-UK and EPSRC Cancer Imaging Centre, Institute of Cancer Research and Royal Marsden Hospital, Sutton, Surrey, United Kingdom; ²CR-UK and EPSRC Cancer Imaging Centre, Institute of Cancer Research and Royal Marsden Hospital, United Kingdom

A major challenge for implementing diffusion-weighted magnetic resonance imaging (DW-MRI) as tumour response biomarker in multicentre clinical trials is in measuring changes in individual patients reliably and reproducibly. The delineation of the region of interest (ROI) has a great impact on final reproducibility. This study compares ADC values obtained using in-house computer ROI drawing software (Diffusion View) to segment tumour with those obtained from a manual drawing technique. Computer generated ROIs have the advantage of less variability, operator independence and significant time saving and support the feasibility of use of automated DWI measurements in clinical trials.

Body Diffusion

Hall B Monday 14:00-16:00 Computer 101

14:00 4710. Liver DTI of Obese Insulin Resistant Subjects with Fatty Liver Disease

Lan Lu¹, Jacob M. Haus², John P. Kirwan², Chris A. Flask<sup>1,3

1</sup>Department of Radiology, Case Western Reserve University, Cleveland, OH, United States; ²Department of Pathobiology, Cleveland Clinic, Cleveland, OH, United States; ³Department of Biomedical Engineering, Case Western Reserve University, Cleveland, OH, United States

Recent studies suggest diffusion-weighted (DWI) and diffusion tensor MRI (DTI) as a promising tool in studying hepatic diseases including hepatitis and cirrhosis. Similarly, Non-Alcoholic Fatty Liver Disease (NAFLD) progresses from fatty liver (steatosis) to fibrosis (steatohepatitis), but diffusion MRI studies on NAFLD subjects are limited. In this study, we used DTI to investigate the diffusion parameters in obese, insulin resistant adults with fatty liver in comparison with healthy volunteers. Our results demonstrate diffusional changes indicative of hepatic adipocyte accumulation and not fibrosis. Therefore, hepatic lipids can complicate the utility of DWI/DTI in the study hepatic fibrosis associated with NAFLD/NASH.

14:30 4711. The Reproducibility of Apparent Diffusion Coefficient Measurement in Liver of Healthy Volunteers at 3.0T and 1.5T Diffusion Weighted Imaging

Jiayin Gao¹, Jinning Li¹, Zhenghan Yang¹, Yuan Fu¹, Liang Xu¹, Min Chen¹, Cheng Zhou<sup>1

1</sup>Radiolgical Department, Beijing Hospital, Beijing, China

The aim of this study is to evaluate the reproducibility of ADC measurement in liver of healthy volunteers at 1.5T and 3.0T MR scanner. DWI of the liver was performed in 30 healthy volunteers with both 1.5T and 3.0T MR scanner in the same day. The mean ADC value of liver was 1.57¡Á10-3 mm2/s at 1.5T, while 1.35¡Á10-3 mm2/s at 3.0T (P<0.001). The result indicates that there is significant variability of ADC measurement in hepatic parenchyma at between 1.5T and 3.0T.

15:00 4712. Application of Diffusion Weighted Proton MRI for Evaluating Molecular Diffusion of Water and Tissue Perfusion in Diethylnitrosamine Induced Rat Liver Fibrosis

Beena George¹, Andriy Babsky¹, George E. Sandusky², Navin Bansal<sup>1

1</sup>Radiology, Indiana University, Indianapolis, IN, United States; ²Pathology and Laboratory Medicine, Indiana University, Indianapolis, IN, United States

The effect of hepatic fibrosis produced by diethylnitrosamine (DEN) was examined by fat and water MRI and diffusion weighted (DW) 1H MRI, separating water molecular diffusion and tissue perfusion. DEN caused an increase in liver water MRI signal intensity but no change in fat content. Fast apparent diffusion coefficient (ADC), which represents perfusion, was lower in DEN treated fibrotic livers compared to control livers. Slow ADC, which represents molecular diffusion, was same in both groups. Measurement of perfusion by DW 1H MRI may be a useful marker for diagnosis of liver fibrosis.

15:30 4713. Biexponential Study in Diffusion Weighted Imaging of Liver Tumours

Jean-Luc Daire¹, Ralph Sinkus², Mathilde Wagner¹, Nathalie Haddad¹, Valérie Vilgrain¹, Bernard Van Beers<sup>1

1</sup>CRB3 Centre de Recherches Biomédicales Bichat-Beaujon, INSERM U773, Paris, France, Metropolitan; ²Institut Langevin, ESPCI, Paris, 75005, France, Metropolitan

The aim of this project was to prospectively evaluate a DW MR imaging sequence combined with parallel acquisition to allow the calculation of pure molecular-based (D) and perfusion related (D*, f) diffusion parameters based on IVIM theory, in liver lesions according to their enhancement behaviour on 3D gradient-echo contrast-enhanced MR sequences. We have shown that D* and f values can reflect liver lesions perfusion since hypervascular liver lesions had significantly greater D* and f values than intermediate and non hypervascular lesions. A biexponential approach might be useful for the characterization of liver lesions and the assessment of tumour response.

Tuesday 13:30-15:30 Computer 101

13:30 4714. Value of Diffusion-Weighted MR Imaging as an Early Surrogate Parameter for the Response of Colorectal Metastases to Interstitial 192Ir-High-Dose-Rate Brachytherapy

Oliver Dudeck¹, Christian Wybranski¹, Martin Zeile¹, David Löwenthal¹, Frank Fischbach¹, Maciej Pech¹, Gero Wieners¹, Jens Ricke<sup>1

1</sup>Radiology and Nuclear Medicine, Otto-von-Guericke University, Magdeburg, Sachsen-Anhalt, Germany

Forty colorectal liver metastases in 30 patients were evaluated with MRI including diffusion-weighted imaging (DWI) immediately before, 2 days after and 90 days following CT- and MR-guided brachytherapy. Tumor diameter (TD) and ADC were evaluated by two radiologists. On early postprocedural MRI, mean TD sligthly incresed while mean ADC decreased significantly (p<0,001). On follow-up MRI, a decrease in mean TD correlated with an increase of mean ADC (p<0,001 both; r=-0,565). In conclusion, changes in ADC can be assessed as soon as 2 days following brachytherapy. Early ADC decrease most likely reflected cell swelling, late increase reduction of tumor cell density.

14:00 4715. Diffusion Weighted Imaging in Differentiating Malignant from Benign Intraductal Papillary-Mucinous Neoplasm of the Pancreas

Atsushi Nakamoto¹, Tonsok Kim¹, Masatoshi Hori¹, Hiromitsu Onishi¹, Takahiro Tsuboyama¹, Mitsuaki Tatsumi¹, Noboru Maeda¹, Hiroki Higashihara¹, Keigo Osuga¹, Kaname Tomoda<sup>1

1</sup>Department of Radiology, Osaka University Graduate School of Medicine, Suita, Osaka, Japan

We retrospectively evaluated the diagnostic performance of diffusion-weighted imaging (DWI) and apparent diffusion coefficient (ADC) value for differentiating malignant intraductal papillary-mucinous neoplasm (IPMN) of the pancreas from benign IPMN. The mean minimum ADC value of malignant IPMN was lower than that of benign IPMN (P < 0.001). Mean Az value, sensitivity, specificity and accuracy of MR imaging with DWI were higher than those of MR imaging without DWI, although differences were not significant. We conclude that DWI and measurement of minimum ADC value would be helpful for differentiating malignant IPMN form benign IPMN.

14:30 4716. New Application Screening for Malignant Tumors Using the Lesion to Spinal Cord Ratio (LSR) in Diffusion Weighted Images

Mamoru Takahashi¹, Yasuo Takehara², Takahiro Natsume³, Norihiro Tooyama, Katsutoshi Ichijo, Harumi Sakahara², Atsushi Nozaki<sup>4

1</sup>Seirei Mikatabara General Hospital, Hamamatsu, Shizuoka, Japan; ²Hamamatsu University of Medicine; ³Seirei Hamamatsu General Hospital; ⁴GE Healthcare Japan

Problem: Tumor detection based on apparent diffusion co-efficient (ADC) is not always efficient and accurate on the high b-value diffusion weighted image (DWI). Method: The signal intensity ratio between lesions and spinal cord signal (LSR) were measured on DWI. The LSR was higher in malignant tumors than in other lesions. Tumor screening application gLSR maph was developed based on the signal intensity ratio between the lesions and the spinal cord signal on DWI. Results: With the cut-off of 0.45 for LSR, sensitivity for malignant tumor was 100%. Using the color map, effective and robust tumor screening was feasible on DWI.

15:00 4717. Nodal Volumetric Histographic ADC Changes Associated with Successful Chemotherapy of Adolescent and Childhood Lymphoma

Shonit Punwani¹, Alan Bainbridge¹, Stuart Taylor¹, Steven Daw², Ananth Shankar², Paul Humphries<sup>1

1</sup>Radiology, University College London Hospital, London, United Kingdom; ²Paediatrics, University College London Hospital, London, United Kingdom

Cellular density should decrease following successful chemotherapy of lymphomatous tissue and a corresponding rise of ADC is expected. This study investigates the extent and nature of histographic post treatment ADC changes associated with a successful treatment outcome in childhood and adolescent patients with lymphoma.

Wednesday 13:30-15:30 Computer 101

13:30 4718. Distortion Correction of Body Diffusion Weighted Images Using Three Point-Dixon Method

Saori Mori¹, Isao Muro¹, Hisamoto Moriguchi¹, Tomohiko Horie¹, Masatoshi Honda¹, Tesuo Ogino², Makoto Obara², Yutaka Imai<sup>1

1</sup>Radiology, Tokai university, Isehara, Kanagawa, Japan; ²Philips Healthcare Asia Pacific, Shinagawa, Tokyo, Japan

Significant distortion can often be observed in body diffusion weighted images using single shot EPI. In this study, we demonstrate that a B0 map obtained using three point-Dixon method is useful to correct for distortion. The correlation coefficients of corrected images were increased by 0.02 or greater from those of the uncorrected images. In this method, although prescan is required, it requires about a minute and a B0 map obtained from the prescan can be applied to 50 images. This method is quite useful in practice since a B0 map created using this method enables good distortion correction.

14:00 4719. Motion Induced Signal Non-Uniformity Correction with Asymmetric Bipolar MPG on Liver DWI

Tetsuo Ogino¹, Tomohiko Horie², Isao Muro², Marc Van Cauteren³, taro takahara<sup>4

1</sup>Clinical Science , Philips Electronics Japan, LTD, Minato-ku, Tokyo, Japan; ²Tokai Univ. Hospital, Japan; ³Philips Healthcare Asia Pacific, minato-ku, Tokyo, Japan; ⁴Division of Radiology, Radiotherapy and Nuclear Medicine (RRN), University Medical Center (UMC) Utrecht, Utrecht, Netherlands

Respiratory and Cardiac motion affects DWI IQ on liver. A voxel deformation due to motion under Motion Probing Gradient(MPG) induces spin phase dispersion in a voxel and results signal loss. By replacing conventional MPG with a pair of bipolar gradient, the signal loss is reduced significantly. However, minimum TE and TR are prolonged due to low efficiency of such type of MPG. We propose a novel MPG method which is formed as hybrid of conventional MPG and bipolar MPG. Image quality of liver diffusion is significantly improved with slight scan time prolongation from conventional respiratory triggered DWI.

14:30 4720. Comparison of Simulation-Based and Measurement-Based RF Shimming for Whole-Body MRI at 7 Tesla

Andreas K. Bitz^1,2, Irina Brote^1,2, Stephan Orzada^1,2, Oliver Kraff^1,2, Stefan Maderwald^1,2, Harald H. Quick³, Klaus Solbach⁴, Achim Bahr⁵, Hans-Peter Fautz⁶, Franz Schmitt⁶, Mark E. Ladd<sup>1,2

1</sup>Erwin L. Hahn Institute for MRI, University Duisburg-Essen, Essen, Germany; ²Department of Diagnostic and Interventional Radiology and Neuroradiology, University Hospital Essen, Essen, Germany; ³Institute for Medical Physics, Friedrich-Alexander-University, Erlangen-Nuernberg, Germany; ⁴High Frequency Technique, University Duisburg-Essen, Duisburg, Germany; ⁵IMST GmbH, Kamp-Lintfort, Germany; ⁶Siemens Healthcare Sector, Erlangen, Germany

The aim of this study was to compare RF shimming procedures based on RF simulations and measured B1+ maps for different ROIs in the thorax and abdomen of volunteers with varying physique. Simulation-based RF shims were computed by use of heterogeneous body models. In vivo B1+ mapping was performed during breath hold by utilization of a pre-saturation turboFLASH. ROIs in the heart, liver, and kidney were considered. Appropriate simulation-based shims could be derived which function in the majority of considered volunteers. If the numerical body model inadequately describes the subject’s body, measured shims should be preferred to achieve higher image quality.

15:00 4721. Combinatorial Fat Suppression for Diffusion Weighted Imaging at 3.0T

Matthew David Blackledge¹, David Higgins², Dow-Mu Koh¹, Nandita M. deSouza¹, Martin O. Leach¹, David J. Collins<sup>1

1</sup>CR-UK and EPSRC Cancer Imaging Centre, Institute of Cancer Research and Royal Marsden Hospital, Sutton, Surrey, United Kingdom; ²Philips Healthcare, Guildford, Surrey, United Kingdom

A variety of fat suppression techniques including STIR, SPIR, SPAIR and SSGR used alone or in combination are investigated for use in large field of view Diffusion Weighted Imaging at 3.0T. We conclude that a combinatorial approach improves results. A combination of STIR and SPIR works well and can be further improved used in combination with SSGR. All methods would benefit from improved shimming over large field of view to reduce off-resonance attenuation effects.