Ruiqi Geng^1,2, Collin J. Buelo^1,2, Mahalakshmi Sundaresan³, Jitka Starekova¹, Nikolaos Panagiotopoulos^1,4, Thekla Helene Oechtering^1,4, Edward M. Lawrence¹, Marcin Ignaciuk¹, Scott B. Reeder^1,2,5,6,7, and Diego Hernando^1,2,3,7
1Radiology, University of Wisconsin-Madison, Madison, WI, United States, ²Medical Physics, University of Wisconsin-Madison, Madison, WI, United States, ³Electrical and Computer Engineering, University of Wisconsin-Madison, Madison, WI, United States, ⁴Radiology and Nuclear Medicine, Universität zu Lübeck, Lübeck, Germany, ⁵Medicine, University of Wisconsin-Madison, Madison, WI, United States, ⁶Emergency Medicine, University of Wisconsin-Madison, Madison, WI, United States, ⁷Biomedical Engineering, University of Wisconsin-Madison, Madison, WI, United States

An automated AI-based method for liver image prescription from a localizer was recently proposed. In this work, this AI method was further evaluated in a larger retrospective patient cohort (1,039 patients for training/testing), across pathologies, field strengths, and against radiologists’ inter-reader reproducibility performance. AI-based 3D axial prescription achieved a S/I shift of <2.3 cm compared to manual prescription for 99.5% of test dataset. The AI method performed well across all sub-cohorts and better in 3D axial prescription than radiologists’ inter-reader reproducibility performance. The AI method was successfully implemented on a clinical MR system and showed robust performance across localizer sequences.

Fasil Gadjimuradov^1,2, Thomas Benkert², Marcel Dominik Nickel², and Andreas Maier^1
1Pattern Recognition Lab, Department of Computer Science, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany, ²MR Application Predevelopment, Siemens Healthcare GmbH, Erlangen, Germany

Signal-dropouts caused by cardiac motion are frequently observed artifacts in diffusion-weighted imaging (DWI) of the liver. When several repetitions of a given slice are affected, uniform averaging results in locally reduced liver signal. This work proposes an adaptive weighted averaging of repetitions to locally suppress signal-dropouts. Therefore, weight maps are estimated by an algorithm which computes robust patch statistics under the guidance of a learned classifier which marks corrupted repetitions. In comparison to other methods, the proposed approach enables more homogeneous liver signal and less biased quantitative maps while sacrificing little signal-to-noise ratio (SNR).

Gavin Hamilton¹, Alexandra N Schlein¹, Walter C Henderson¹, Danielle N Batakis¹, Lael K Ceriani¹, Ashley L Louie¹, Michael S Middleton¹, Yesenia Covarrubias¹, Kathryn J Fowler¹, and Claude B Sirlin^1
1Liver Imaging Group, Department of Radiology, University of California San Diego, San Diego, CA, United States

Using MRS, we examined whether the assumed fat spectral model (6-, or 9-peak) altered liver R2* estimation in vivo.  We found that liver R2* estimation was model-independent, and liver R2*was not significantly different from water R2*  (p = 0.07). However, fat R2* values estimated for these models were different (p < 0.0001), suggesting that fat R2* estimation in high-fat tissues may be model dependent.

Meryem KHALFALLAH¹, Sabrina Doblas¹, Felicia Julea¹, Catherine Postic², Dominique Valla³, Valérie Paradis^3,4, Philippe Garteiser¹, and Bernard Van Beers^1,5
1Université de Paris, Laboratory of Imaging Biomarkers, Center of Research on Inflammation, UMR 1149, Inserm, F-75018 Paris, France, Paris, France, ²Université de Paris, Institut Cochin, CNRS, INSERM, Paris, France., Paris, France, ³Service d’hépatologie et UMR1149, Hôpital Beaujon, Clichy-la-Garenne ; APHP, Université de Paris et Inserm, Clichy-la-Garenne, France, ⁴Diderot, CNRS, Centre de Recherche sur l'Inflammation (CRI), Paris, F-75890, France., Paris, France, ⁵Department of Radiology, AP-HP, Beaujon University Hospital Paris Nord, F-92110 Clichy, France, Clichy, France

Nonalcoholic steatohepatitis (NASH) and significant metabolic liver disease (MLD) are progressive forms of nonalcoholic fatty liver disease (NAFLD). MR elastography parameters have been shown to be useful for assessing NAFLD/NASH and liver inflammation. To assess the role of viscoelastic parameters in diagnosing NASH and significant MLD, MRE hepatic examinations were performed in mice with NAFLD. Our results showed that the diagnostic performance of MR elastography was higher in significant MLD than in NASH. This may be explained by inflammation and fibrosis occurring in indeterminate NASH cases included in MLD, and by poor performance of MR elastography to detect hepatocyte ballooning.

Francesco Grussu¹, Kinga Bernatowicz¹, Ignasi Barba², Irene Casanova-Salas³, Natalia Castro³, Marco Palombo⁴, Sara Simonetti^3,5, Juan Francisco Corral^6,7, Marta Vidorreta⁸, Xavier Merino^6,7, Richard Mast^6,7, Núria Roson^6,7, Manuel Escobar Amores^6,7, Nahúm Calvo‐Malvar^6,9, Josep R. Garcia-Bennett⁹, Rodrigo A. Toledo¹⁰, Joaquin Mateo³, Paolo Nuciforo⁵, Elena Garralda¹¹, and Raquel Perez-Lopez^1,7
1Radiomics Group, Vall d'Hebron Institute of Oncology, Vall d'Hebron Barcelona Hospital Campus, Barcelona, Spain, ²NMR Lab, Vall d'Hebron Institute of Oncology, Vall d'Hebron Barcelona Hospital Campus, Barcelona, Spain, ³Prostate Cancer Translational Research Group, Vall d'Hebron Institute of Oncology, Vall d'Hebron Barcelona Hospital Campus, Barcelona, Spain, ⁴Cardiff University Brain Research Imaging Centre, Cardiff University, Cardiff, United Kingdom, ⁵Molecular Oncology Group, Vall d'Hebron Institute of Oncology, Vall d'Hebron Barcelona Hospital Campus, Barcelona, Spain, ⁶Institut de Diagnòstic per la Imatge (IDI), Catalonia, Spain, ⁷Department of Radiology, Hospital Universitari Vall d'Hebron, Barcelona, Spain, ⁸Siemens Healthineers, Madrid, Spain, ⁹Hospital Universitari de Bellvitge, L'Hospitalet de Llobregat, Spain, ¹⁰Gastrointestinal and Endocrine Tumors Group, Vall d'Hebron Institute of Oncology, Vall d'Hebron Barcelona Hospital Campus, Barcelona, Spain, ¹¹Early Clinical Drug Development Group, Vall d'Hebron Institute of Oncology, Vall d'Hebron Barcelona Hospital Campus, Barcelona, Spain

“Diffusion-Relaxation Hepatic Imaging via Generalised Assessment of DiffusiOn Simulations” (DR-HIGADOS) is a recent liver diffusion MRI method for intra-cellular fraction, cell size and cellularity measurement. Here we compare DR-HIGADOS metrics to counterparts from histology in two data sets: i) preclinical 9.4T MRI of formalin-fixed patient-derived xenograft mouse livers, with co-localised histological sections; ii) clinical 1.5T in vivo MRI of metastatic and primary liver cancer patients, with co-localised biopsies. Results confirm that DR-HIGADOS cell size and cellularity are histologically meaningful, paving the way to the application of the new technique in clinical studies.

Yavuz Muslu^1,2, Ty A. Cashen³, Sagar Mandava⁴, and Scott B. Reeder^1,2,5,6,7
1Department of Biomedical Engineering, University of Wisconsin-Madison, Madison, WI, United States, ²Department of Radiology, University of Wisconsin-Madison, Madison, WI, United States, ³GE Healthcare, Waukesha, WI, United States, ⁴GE Healthcare, Atlanta, GA, United States, ⁵Department of Medical Physics, University of Wisconsin-Madison, Madison, WI, United States, ⁶Department of Medicine, University of Wisconsin-Madison, Madison, WI, United States, ⁷Department of Emergency Medicine, University of Wisconsin-Madison, Madison, WI, United States

Gadoxetic acid (GA)-enhanced hepatobiliary phase T₁-weighted (HBP-T₁w)-MRI is widely used in the clinical setting for detection of focal liver lesions (FLLs). However, complete characterization of liver lesions also requires T₂w, diffusion weighted (DWI) and dynamic contrast enhanced (DCE) imaging. These methods often have inferior image spatial resolution compared to GA-enhanced HBP-T₁w and may be unable to resolve small (<1cm) lesions visible only on HBP-T₁w. In this work, we propose a novel phase sensitive inversion recovery (PSIR) T₁w-MRI in combination with 3D radial stack-of-stars imaging and model-based water/fat separation for improved detection and characterization of small FLLs.

Christopher R Bradley^1,2, Deirdre Mcgrath^1,2, Eleanor F Cox^1,2, Naaventhan Palaniyappan², Robert Scott², Indra N Guha², Guruprasad P Aithal², and Susan T Francis^1,2
1Sir Peter Mansfield Imaging Centre, University of Nottingham, Nottingham, United Kingdom, ²NIHR Biomedical Research Centre, Nottingham University Hospitals NHS Trust, Nottingham, United Kingdom

We assess liver stiffness measured using magnetic resonance elastography (MRE) and longitudinal relaxation time T₁ in three groups:  healthy volunteers, non-alcoholic fatty liver disease patients and compensated cirrhosis patients. MRE liver stiffness was measured using the Quantitative Imaging Biomarker Alliance (QIBA) recommendation. T₁ longitudinal relaxation time of the liver was measured using a respiratory triggered inversion recovery fat-suppressed spin-echo echo planar imaging scheme. A positive correlation between Liver tissue T₁ and Liver stiffness (R=0.70, p<0.0001) across the 3 groups was observed. Superior mesenteric artery flow also correlates with liver stiffness (R=0.62, p<0.002) suggesting worsening hyperdynamic circulation with progressive fibrosis.

Sherry Huang¹, Yong Chen², Reid Bolding³, Leonardo Kayat Bittencourt^2,4, Mark A Griswold², and Rasim Boyacioglu^2
1Biomedical Engineering, Case Western Reserve University, Cleveland, OH, United States, ²Radiology, Case Western Reserve University, Cleveland, OH, United States, ³Physics, Case Western Reserve University, Cleveland, OH, United States, ⁴Radiology, University Hospitals Cleveland Medical Center, Cleveland, OH, United States

This study presents a Pilot Tone (PT) based free-breathing technique for two-dimensional simultaneous quantification of T1, T2, T2*, fat fraction (FF), and water fraction (WF) using quadratic RF phase Magnetic Resonance Fingerprinting (qRF-MRF). We report the quantitative comparison of a cohort of 10 healthy subjects between free-breathing and breath-hold qRF-MRF as well as the comparison to standard clinical protocols. Free-breathing results are comparable to both breath-hold and standard clinical protocols (p > 0.05), indicating the stability and reproducibility of the method.

Manuel Taso¹, Donald C Simonson², Lauren Phung¹, Colleen McGrath¹, Stephanie Waldman¹, Fotini Papadopoulou¹, and David C Alsop^1
1Division of MRI Research, department of Radiology, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, United States, ²Division of Endocrinology, Diabetes and Hypertension, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, United States

Loss of β-cell function and mass are hallmarks of type-1 and type-2 diabetes. Non-invasive imaging of endocrine pancreatic function would provide valuable insights into the degree and temporal patterns of endocrine dysfunction. For this purpose, Arterial Spin Labeling (ASL) is a uniquely suited non-invasive technique. We therefore conducted a dynamic ASL study during a hyperglycemic clamp to explore the dynamics of insulin secretion and pancreatic blood-flow.

This study showed that hyperglycemia and arginine-stimulated insulin secretion causes an increase in pancreatic perfusion and that the rate of insulin concentration change seems to be driving perfusion changes rather than absolute concentration.

Chengyue Wu¹, Thomas E Yankeelov^1,2,3,4,5,6, and John Virostko^1,2,3,4
1Oden Institute for Computational Engineering and Sciences, The University of Texas at Austin, Austin, TX, United States, ²Livestrong Cancer Institutes, The University of Texas at Austin, Austin, TX, United States, ³Department of Diagnostic Medicine, The University of Texas at Austin, Austin, TX, United States, ⁴Department of Oncology, The University of Texas at Austin, Austin, TX, United States, ⁵Department of Biomedical Engineering, The University of Texas at Austin, Austin, TX, United States, ⁶Department of Imaging Physics, The University of Texas MD Anderson Cancer Center, Houston, TX, United States

Monitoring of the pancreatic islet is needed to assess individuals at risk for developing diabetes and facilitate development of islet-directed therapies. Pancreatic islets may be detectable due to their high vascularity. We imaged pancreas perfusion using both whole field-of-view and reduced field-of-view Intravoxel Incoherent Motion (IVIM) and found that the reduced field-of-view technique achieved similar image quality in half the time. We then performed a reduced field-of-view IVIM acquisition on four healthy controls. We found that pancreas perfusion increased dynamically in response to oral glucose challenge, exhibiting a biphasic increase that parallels the time course of insulin secretion.

Jan Weis¹ and Olle Korsgren^2
1Department of Medical Physics, Uppsala University Hospital, Uppsala, Sweden, ²Department of Immunology, Genetics and Pathology, Uppsala University, Uppsala, Sweden

Phosphorus (³¹P) and proton (¹H) magnetic resonance spectroscopy (MRS) are the methods of choice in the assessment of the pancreas graft quality before organ or islet of Langerhans transplantation. During the transport and MR scanning hypothermic storage (4±2 ^oC) has to be maintained. The aim of this study was to investigate if it is possible to measure pancreas graft temperature in a MR scanner using ¹H-MRS and temperature constants obtained by the calibration experiments with the water-vegetable oil phantom. The present study has shown that ¹H-MRS is able to measure the graft temperature during MR scanning.

Takahiro Matsuyama¹, Yoshiharu Ohno^1,2, Kaori Yamamoto³, Masato Ikedo³, Masao Yui³, Saki Takeda⁴, Akiyoshi Iwase⁴, Yuka Oshima¹, Nayu Hamabuchi¹, Satomu Hanamatsu¹, Yuki Obama¹, Hiroyuki Nagata¹, Takahiro Ueda¹, Hirotaka Ikeda¹, Kazuhiro Murayama², and Hiroshi Toyama^1
1Radiology, Fujita Health University School of Medicine, Toyoake, Japan, ²Joint Research Laboratory of Advanced Medical Imaging, Fujita Health University School of Medicine, Toyoake, Japan, ³Canon Medical Systems Corporation, Otawara, Japan, ⁴Radiology, Fujita Health University Hospital, Toyoake, Japan

We hypothesized that deep learning reconstruction (DLR) and multiple k-space data by means of each of the repetition time (TR) techniques (Fast 3D mode multiple: Fast 3Dm) are more useful than parallel imaging (PI) and compressed sensing (CS) for shortening acquisition time and improving image quality and IPMN evaluation capability on 3D MRCP.  The purpose of this study was thus to compare the utility of DLR used for PI, Fast 3Dm and CS for improvement of acquisition time, image quality and IPMN evaluation capability on 3D MRCP for patients with IPMN.