Luc Beuzit¹, Pierre-Antoine Eliat², Elise Bannier^1,3, Jean-Christophe Ferré^1,3, Yves Gandon¹, Vanessa Brun¹, and Hervé Saint-Jalmes^4,5
1Radiology, CHU Rennes, Rennes, France, ²PRISM-Biosit CNRS UMS 3480, INSERM UMS 018, University of Rennes I, Rennes, France, ³Neurinfo MR imaging platform, University of Rennes I, Rennes, France, ⁴Radiology, Eugène Marquis Cancer Institute, Rennes, France, ⁵LTSI, UMR 1099, INSERM, University of Rennes I, Rennes, France

Dynamic contrast-enhanced MRI is a promising biomarker for investigating early tumor response, but is subject to numerous factors of variation. The purpose of this study was to assess the potential role of software packages in this variability. Using an original DICOM file processing method, we studied the accuracy of software packages for measuring pharmacokinetic parameters on simulated data. We showed significant errors for all five software packages studied, leading to poor inter-software agreement. A very poor agreement was observed among the software packages when processing clinical data consisting of dynamic contrast-enhanced MR images of rectal tumors.

Stefan Hindel¹, Anika Sauerbrey¹, Marc Maaß², and Lutz Lüdemann^1
1Strahlenklinik und Poliklinik, Universitätsklinikum Essen, Essen, North Rhine-Westphalia, Germany, ²Evangelisches Krankenhaus Wesel GmbH, North Rhine-Westphalia, Germany

We validated the feasibility of absolute regional perfusion quantification by dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) with k-space shared sampling and a blood pool contrast agent in combination with a two-compartment tracer kinetic model in low-perfusion tissue. Seven female pigs were investigated. An ultrasonic Doppler probe was attached to the right femoral artery to determine the total flow in the hind leg musculature. The overall flow of the hind leg muscles, as measured by the ultrasound probe, highly correlated with the total flow from the MRI measurement, r = 0.89 and P=10^(-7).

Xia Zhao^1,2, Yiqun Xue^1,2, Mark Rosen², Hyunseon Kang³, Ramesh Rengan⁴, and Heekwon Song^1,2
1Laboratory for Structural NMR Imaging, University of Pennsylvania, Philadelphia, PA, United States, ²Department of Radiology, Hospital of University of Pennsylvania, Philadelphia, PA, United States, ³MD Anderson Cancer Center, University of Texas, Houston, TX, United States, ⁴Department of Radiation Oncology, University of Washington School of Medicine, Seattle, WA, United States

In this work, we investigate the effects of temporal resolution on measured tumor perfusion parameters and also its effects on repeatability on a test-retest dataset acquired in vivo in patients with lung tumor. We utilize interleaved radial acquisition (golden angle view angle increment) along with KWIC reconstruction which enables different temporal resolutions to be chosen retrospectively during image reconstruction. Our results demonstrate that repeatability is relatively constant up to about 13 sec frame rate (CV ~ 11%). In agreement with previous simulation studies, K^trans is increasingly underestimated at reduced temporal resolutions.

Yash Vardhan Tiwari^1,2, Qiang Shen¹, Zhao Jiang¹, Wei Li¹, Justin Long^1,2, Chenling Fang^1,2, and Timothy Duong^1
1Research Imaging Institute, UT Health Science Center, San Antonio, Texas, United States, ²Biomedical Engineering, UT, San Antonio, Texas, United States

Blood-brain-barrier (BBB) dysfunction has been implicated in a number of neurological disorders, such as multiple sclerosis, stroke and cancer. BBB integrity can be assessed by measuring the water exchange rate across the BBB (Kw) – defined as PS/Vc where PS is permeability surface area product and Vc is capillary distribution volume of water. Diffusion-weighted arterial spin labeling (DW-ASL) was recently proposed to measure Kw without using an exogenous contrast agent in humans1. However, this approach has not been validated. The goal of this study was to compare Kw measurements (DW-ASL) with Ktrans (DCE MRI) and histological staining (Evan’s blue) in the same rats. Measurements were also made before and after mannitol administration to break the BBB.

Tracy Ssali^1,2, Udunna C Anazodo^1,2, Mahsa Shokouhi¹, Bradley J MacIntosh³, and Keith St Lawrence^1,2
1Laswon Health Research Institute, London, Ontario, Canada, ²University of Western Ontario, London, Ontario, Canada, ³Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada

It is known that knowledge of the arterial transit time (ATT) is essential for CBF quantification; however, typically ATT imaging is time consuming. The aim of this study was to assess intra- and inter-subject reproducibility of ATT. Across sessions and subjects, the voxel-wise GM CV were 6.0% and 10.0%, respectively. Variance in mean GM ATT inter-subject was 44.7ms (CV=5.4%) and 5.9ms (CV=0.7%) inter-session. The low variance demonstrates the stability of ATT measurements in young healthy individuals. Next, would be to assess reproducibility of ATT in older subject groups and patient populations with cerebral pathology where perfusion characteristics have greater variability.

Clement Stephan Debacker^1,2, Jan M Warnking^1,3, Sacha Koehler², Jerome Voiron², and Emmanuel L Barbier^1,3
1GIN, Univ. Grenoble Alpes, Grenoble, France, ²Bruker BioSpin MRI, Ettlingen, Germany, ³U836, INSERM, Grenoble, France

The purpose of this study is to evaluate the reproducibility across magnetic field of cerebral blood flow (CBF) quantification using Arterial Spin Labeling (ASL) methods. Three ASL methods (pulsed ASL (PASL), continuous ASL (CASL), and pseudo-continuous ASL (pCASL)) were evaluated at 4.7, 7, and 11.7T in 43 rats. Among the three techniques evaluated, pCASL appears as the most reproducible method across magnetic fields.

Lirong Yan¹, Collin Liu², Robert Smith¹, Mayank Jog¹, Kate Krasileva¹, Cheng Li³, Michael Langham³, and Danny JJ Wang^1
1Neurology, University of California Los Angeles, Los Angeles, CA, United States, ²University of Southern California, CA, United States, ³University of Pennsylvania, Philadelphia, PA, United States

The purpose of this study was to investigate the relationship between intracranial vascular compliance and aortic pulse wave velocity (PWV) in a cohort of subjects with mixed vascular risks. We found there was significant negative correlation between intracranial VC and aortic PWV. Subjects with high vascular risks showed reduced intracranial VC and elevated aortic PWV compared to subjects with low vascular risks. These findings suggest that changes in vascular compliance typically extend from aorta to intracranial vessels.

Natalie M Wiseman¹, Meng Li², Mahmoud Zeydabadinezhad³, Jessy Mouannes-Srour³, Yongquan Ye², E. Mark Haacke^2,3, and Zhifeng Kou^2,3
1Department of Psychiatry and Behavioral Neurosciences, Wayne State University School of Medicine, Detroit, MI, United States, ²Department of Radiology, Wayne State University School of Medicine, Detroit, MI, United States, ³Department of Biomedical Engineering, Wayne State University, Detroit, MI, United States

DSC-PWI suffers from blooming, clipping, and saturation effects in large vessels, which make arterial input function (AIF) determination unreliable. To combat this, we used a dual-injection method in which 1/6 of the contrast dose is used to determine the AIF and the remaining 5/6 is used to visualize perfusion in brain tissue. We compared these CBF measurements to pseudo-continuous ASL (pCASL) CBF measurements. The correlation coefficients of 0.85 and 0.63 show good correlation in large ROIs and slightly less good correlation in the small ROIs, respectively. In conclusion, the dual-injection DSC-PWI method provides a reliable, high resolution measurement of CBF.

Ryan J Bosca¹ and Edward F Jackson^1
1Medical Physics, University of Wisconsin - Madison, Madison, WI, United States

Increasing interest in and use of quantitative imaging biomarkers in clinical research and clinical practice necessitates careful assessment and mitigation of bias and variance associated with the acquisition and image analysis techniques, especially in the context of spatially heterogeneous processes. In this work, we report an extensible methodology for creating an anthropomorphic digital reference object (DRO) with a synthetic tumor to evaluate quantitative imaging algorithms using a realistic data analysis scenario with known “ground truth”. Such a realistic DRO facilitates quantitative imaging biomarker algorithm comparisons and validation.

Ina Nora Kompan^1,2, Klaus Eickel^3,4, Federico von Samson-Himmelstjerna^1,5, Benjamin Richard Knowles⁶, and Matthias Guenther^1,2
1Fraunhofer MEVIS, Bremen, Bremen, Germany, ²mediri GmbH, Heidelberg, Baden-Württemberg, Germany, ³Fraunhofer MEVIS, Bremen, Germany,⁴Universitätsklinikum Essen, Essen, Nordrhein-Westfalen, Germany, ⁵Charité, Berlin, Germany, ⁶Universitätsklinikum Freiburg, Freiburg, Baden-Württemberg, Germany

For optimization and validation of new dynamic contrast-enhanced MRI sequences, a simple and cheap perfusion phantom is developed. Contrast agent is injected via a hose into a sponge and is washed out by a constant water flow. The phantom can be quantitatively described using a gamma-variate function, providing heterogeneous parameter maps. The phantom experiment is relatively well reproducible.