View Presentation Video

Tobias Klein^1,2, Thomas Gladytz¹, Jason M. Millward¹, Kathleen Cantow³, Luis Hummel³, Erdmann Seeliger³, Sonia Waiczies¹, Christoph Lippert^2,4, and Thoralf Niendorf^1,5
1Berlin Ultrahigh Field Facility (B.U.F.F.), Max Delbrück Center for Molecular Medicine, Berlin, Germany, ²Digital Health - Machine Learning Research Group, Hasso Plattner Institute for Digital Engineering, University of Potsdam, Potsdam, Germany, ³Institute of Translational Physiology, Charité - Universitätsmedizin, Berlin, Germany, ⁴Hasso Plattner Institute for Digital Health, Icahn School of Medicine at Mount Sinai, New York City, NY, United States, ⁵Experimental and Clinical Research Center, a joint cooperation between the Charité Medical Faculty and the Max Delbrück Center for Molecular Medicine, Berlin, Germany

Keywords: Kidney, Preclinical

Deep learning algorithms enable fast kidney segmentation, which is crucial to establish renal size as a (pre)clinical biomarker for renal diseases. Tackling this challenge, a novel deep dilated U-Net (DDU-Net) was trained, validated, and tested on preclinical ground truth data, benchmarked on simulated data against an analytical model, and applied to longitudinal in vivo MRI scans acquired in rats, with pathophysiological interventions mimicking clinically realistic scenarios. Our DDU-Net reached a Dice score of 0.98 on the ground truth, outperformed the analytical approach, and facilitated rapid detection of acute changes in kidney size upon acute pathophysiological interventions.

View Presentation Video

Chang Ni¹, Xin Mu¹, Yanbin Li², Haikun Qi^3,4, and Jeff L. Zhang^1
1Vascular and Physiologic Imaging Research (VPIR) Lab, School of Biomedical Engineering, ShanghaiTech University, Shanghai, China, ²Central Research Institute,UIH Group, Shanghai, China, ³School of Biomedical Engineering, ShanghaiTech University, Shanghai, China, ⁴Shanghai Clinical Research and Trial Center, Shanghai, China

Keywords: Kidney, Segmentation

Segmentation of renal cortex in MR images is important but challenging. In this study, we proposed to pre-train a ResUNet model with CT images and to use an automatic method for labeling renal cortex for the training data. Such method with transfer learning and automatic labeling performed well in segmenting renal cortex in MR images, with a DICE similarity of 0.85 and volume error of 14%±5%. The proposed method would make labeling of renal cortex for training dataset much more efficiently, and we further confirm the power of transfer learning technique in segmenting renal MR images.

Ye Ju¹, Yue Wang¹, Nan Wang¹, and Ailian Liu^1
1the First Affiliated Hospital of Dalian Medical University, Dalian, China

Keywords: Kidney, fMRI

Patients with liver cirrhosis are easily complicated with renal injury, and it is often difficult to find the symptoms of early renal injury in time, which eventually leads to poor prognosis. T2 mapping, as a quantitative MRI technique, can provide quantitative information about the content and composition of water and collagen fibers [1]. The results of this study show that T2 mapping can effectively evaluate the renal damage of liver cirrhosis, so T2 mapping is expected to be an effective way to quantitatively and non-invasively evaluate the renal damage of liver cirrhosis. 

Yue Wang¹, Ye Ju¹, Nan Wang¹, and Ailian Liu^1
1the First Affiliated Hospital of Dalian Medical University, Dalian, China

Keywords: Kidney, Liver

Patients with liver cirrhosis are easy to be complicated with renal function injury, and symptoms of early renal injury are hidden and often difficult to be detected in time, which may ultimately lead to poor prognosis. mDixon-Quant imaging, a novel fat quantitative MRI technique was used to evaluate the degree of renal function injury in patients with cirrhosis. Results showed that mDixon-Quant imaging can be an effective method for noninvasive diagnosis and early assessment of renal function injury in liver cirrhosis.

View Presentation Video

Haitham Al-Mubarak¹, Octavia Bane^1,2, Paul Kennedy¹, Philip Robson¹, Jordan Cuevas^1,2, Arthi Reddy², Kristina Prachanronarong², Kirolos Meilika³, Amir Horowitz⁴, Bernd Kuhn⁵, Steven Sourbron⁶, Ketan Badani³, Bachir Taouli^1,2, and Sara Lewis^1,2
1BioMedical Engineering and Imaging Institute, Icahn School of Medicine at Mount Sinai, New York, NY, United States, ²Department of Diagnostic, Molecular and Interventional Radiology, Icahn School of Medicine at Mount Sinai, New York, NY, United States, ³Department of Urology, Icahn School of Medicine at Mount Sinai, New York, NY, United States, ⁴Precision Immunology Institute/Tisch Cancer Institute, Icahn School of Medicine at Mount Sinai, New York, NY, United States, ⁵Siemens Healthcare, Erlangen, Germany, ⁶University of Sheffield, Sheffield, United Kingdom

Keywords: Cancer, DSC & DCE Perfusion

Functional imaging using dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) and arterial spin-labeling (ASL), which is a non-contrast technique, have been proposed to quantitatively assess perfusion and vascular flow in small renal masses (SRM) and may provide complementary information. In our initial series, we found significant differences in certain DCE-MRI quantitative and semi-quantitative parameters for renal cell carcinoma (RCC) versus benign tumors (Vp, wash-in, wash-out TTP and AUC) and between ccRCC versus non-clear cell (non-cc)RCC (Ve), while no differences existed between groups for ASL. Certain pharmacokinetic parameters derived from DCE-MRI appear promising for SRM characterization

View Presentation Video

Kaixuan Zhao^1,2,3, Sheng Li⁴, Zhigang Wu⁵, Thoralf Niendorf⁶, Wenjian Wang⁴, and Zaiyi Liu^1,3
1Department of Radiology, Guangdong Provincial People’s Hospital, Guangdong Academy of Medical Sciences,Guangzhou 510080, China, Guangzhou, China, ²Guangdong Cardiovascular Institute, Guangdong Provincial People's Hospital, Guangdong Academy of Medical Sciences, Guangzhou 510080, China, Guangzhou, China, ³Guangdong Provincial Key Laboratory of Artificial Intelligence in Medical Image Analysis and Application, Guangdong Provincial People's Hospital, Guangdong Academy of Medical Sciences, Guangzhou 510080, China, Guangzhou, China, ⁴Division of Nephrology, Guangdong Provincial People’s Hospital, Guangdong Academy of Medical Sciences, Guangzhou, China, Guangzhou, China, ⁵Philips Healthcare Beijing Ltd., Shenzhen, China, ⁶Berlin Ultrahigh Field Facility (B.U.F.F.), Max Delbrueck Center for Molecular Medicine in the Helmholtz Association, Berlin, Germany, Berlin, Germany

Keywords: Kidney, Kidney

Diabetic Kidney Disease (DKD) is the leading cause of end-stage renal disease worldwide. Morphological evaluation of renal compartment volume changes and functional evaluation of renal ADC changes provide comprehensive insight into DKD disease progression. This study compares renal compartment volume changes including: 1) parenchyma volume ,2) pelvis volume, 3) whole kidney volume and 4) parenchyma volume percentage and the functional parameter ADC for DKD groups with different CKD staging. Our results demonstrate that cortical/medullary ADC, parenchyma volume and whole kidney volume are correlated with eGFR and suggest that the cortical ADC is a stronger predictor of eGFR in DKD. 

View Presentation Video

Huay Din¹, Christina J. MacAskill², Sree Harsha Tirumani^2,3, Pew-Thian Yap⁴, Mark Griswold^2,3, Chris Flask^2,3, and Yong Chen^1,3
1Case Western Reserve University, Cleveland, OH, United States, ²Department of Radiology, Case Western Reserve University, Cleveland, OH, United States, ³Department of Radiology, University Hospitals Cleveland Medical Center, Cleveland, OH, United States, ⁴Department of Radiology, University of North Carolina at Chapel Hill, Chapel Hill, NC, United States

Keywords: Kidney, Quantitative Imaging, Cancer

In this pilot study, we developed and optimized a spatially-constrained convolutional neural network to accelerate the scan time for 2D kidney Magnetic Resonance Fingerprinting (MRF). Our results suggest that an acceleration factor of 3 can be achieved with the proposed method, which shortens the 2D breath-hold MRF scan from 15 sec to 5 sec. In addition, the deep learning based approach can be applied for T₁ and T₂ quantification of both normal renal tissues and pathologies including renal cell carcinoma.

View Presentation Video

Yuki Makino¹, Naoki Ohno², Johannes M Peeters³, Masami Yoneyama⁴, Yu Ueda⁴, Tosiaki Miyati², Yukihiro Matsuura¹, Toshifumi Gabata⁵, and Satoshi Kobayashi^1,2,5
1Radiology Division, Kanazawa University Hospital, Ishikawa, Japan, ²Faculty of Health Sciences, Institute of Medical, Pharmaceutical and Health Sciences, Kanazawa University, Ishikawa, Japan, ³Philips Healthcare, Best, Netherlands, ⁴Philips Japan, Tokyo, Japan, ⁵Department of Radiology, Kanazawa University Hospital, Ishikawa, Japan

Keywords: Kidney, Diffusion/other diffusion imaging techniques

The clinical application of triexponential intravoxel incoherent motion analysis in the kidney has been limited because the blurring and misregistration of images caused by respiratory motion during acquisition. To solve this issue, we applied retrospective motion-corrected (MC) reconstruction to free-breathing (FB) DWI of the kidney for accurate triexponential diffusion analysis. Our results showed that triexponential analysis using MC FB-DWI can improve the measurement accuracy and repeatability of triexponential diffusion parameters of the kidney.

View Presentation Video

Peter Wawrzyn¹, Ruben Ngnitewe Massa¹, Jamal Gardezi¹, and Andrew Wentland^1,2,3
1Department of Radiology, University of Wisconsin Madison, Madison, WI, United States, ²Department of Biomedical Engineering, University of Wisconsin Madison, Madison, WI, United States, ³Department of Medical Physics, University of Wisconsin Madison, Madison, WI, United States

Keywords: Kidney, Machine Learning/Artificial Intelligence, Bias Field Correction

A deep learning model was developed for the differentiation of benign and malignant solid renal masses. A representative and balanced dataset was used, and model performance was evaluated with and without bias field correction (BFC). The model input contained multiple channels for various MRI-based tissue contrast weightings. The model achieved an accuracy and AUC of 74% and 71% with BFC and 80% and 59% without BFC. This work showed that a convolutional neural network can be trained to differentiate benign from malignant renal masses with a high degree of accuracy and that BFC improves AUC.

Ruirui Ma¹, Min Jia¹, Xia Wang¹, Wei Zhang¹, Na Zhao¹, Gang Tian¹, Chanjuan Yu¹, Xiuzheng Yue², and Yuedong Han^1
1Xi'an GaoXin Hospital, Xi'an, China, ²Philips Healthcare, Beijing, China

Keywords: Kidney, CEST & MT

Respiratory movement is the main reason for upper and middle abdominal organ movement, which significantly affects the quality of MR imaging.This study aims to analyze the influence of renal motion amplitude on renal APT imaging through free breathing (FB), and breath holding (BH) based on multi-phase balanced turbo fast field echo (B-TFE). The results showed that renal motion amplitude was negatively correlated with APTw image quality scores. The renal motion amplitude of bilateral kidneys of BH mode was significantly reduced compared with FB, and the APT image quality of BH mode was better.

View Presentation Video

Jian Zhao¹, Hai-Yi Wang¹, Mei-Feng Wang¹, Huan-Huan Kang¹, Hui-Ping Guo¹, Shao-Peng Zhou¹, and Hui-Yi Ye^1
1Department of Radiology, Chinese PLA General Hospital, Beijing, China

Keywords: Kidney, Diffusion/other diffusion imaging techniques

Problem: Evaluation of VTT consistency through preoperative images has not been reported yet.
Methods: Intravoxel incoherent motion (IVIM)-derived parameters and apparent diffusion coefficient (ADC) were employed to predict VTT consistency of RCC.
Results: Dt of the primary tumor and VTT generated the area under the ROC curve (AUC) values of 0.758 (95%CI: 0.671, 0.832) and 0.712 (95%CI: 0.622, 0.792) for differentiating friable and solid VTT. The AUC value of the model combining Dp and Dt of VTT and Dt of the primary tumor reached 0.886 (95%CI: 0.814, 0.937).
Conclusions: IVIM parameters can effectively predict the VTT consistency of RCC.
 

View Presentation Video

Yu-Wei Hao¹, Hui-Yi Ye¹, and Hai-Yi Wang^1
1Department of Radiology, the first Medical Centre of Chinese PLA General Hospital, Beijing, China

Keywords: Kidney, Kidney, Clear cell likelihood score; Clear cell renal cell carcinoma

Whether T₂-weighted imaging with or without fat suppression technique affects the performance of Clear Cell Likelihood Score v2.0 (ccLS v2.0) for diagnosing ccRCC in SRMs is not known. The final results showed that the tumor signal intensity between T₂-weighted imaging with and without fat suppression was statistically significant (P < 0.001), but diagnostic performance and interobserver agreement of ccLS v2.0 in the diagnosis of ccRCC based on the two imaging techniques was comparable (P>0.05). Fat-suppressed T2-weighted imaging is also applicable to ccLS v2.0. This finding is helpful to expand the clinical application scope of ccLS and increase its clinical universality.

View Presentation Video

Gregory McClanahan^1,2 and Arunark Kolipaka^1,2
1Biomedical Engineering, Ohio State University, Columbus, OH, United States, ²Radiology, Ohio State University, Columbus, OH, United States

Keywords: Kidney, Elastography, MRI, MRE, Diffusion Imaging

Kidney disease and injury can be assessed using tissue stiffness measurements determined by Magnetic Resonance Elastography (MRE), a method that uses MRI. Similarly, diffusion measurements such as apparent diffusion coefficient (ADC) can also provide useful information regarding tissue health. Our goal is to assess the health of the cortex, medulla, and whole kidney by comparing ADC measurements from monopolar and bipolar diffusion gradient schemes against MRE derived stiffness values. Preliminary results indicate a strong correlation between MRE derived stiffness values and monopolar derived ADC measurements and a fair correlation between MRE derived stiffness values and bipolar derived ADC measurements.

View Presentation Video

Hang Su#¹, Gang Zhang#¹, Jingjing Wu¹, Wei Xing¹, Yufu Hu¹, Zhiwei Shen², Ke Jiang², Yuying Sun¹, Xia Zhang¹, and Xianqing Ren*^1
1The First Affiliated Hospital of Henan University of CM, Zhengzhou, China, ²Philips healthcare, Beijing, China

Keywords: Kidney, Kidney

Currently, the therapeutic significance of IVIM-DWI has been examined in numerous kidney illnesses such as transplanted kidney, renal tumor, and renal failure; however, the application in purpura nephritis is uncommon. As a result, the purpose of this study was to investigate the clinical feasibility of IVIM-DWI in differentiating purpura nephritis (HSPN) from allergic purpura (HSP) in children by using D and K values obtained from a small-field DWI sequence with respiratory trigger

Ping Liang¹, Chuou Xu¹, Daoyu Hu¹, and Zhen Li^1
1Radiology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China

Keywords: Kidney, Diffusion/other diffusion imaging techniques

Purpose To explore whether IVIM-DWI and BOLD-MRI can assess the renal function, Oxford classification and prognostic risk stratification of IgAN patients. 

Methods This retrospective study included 46 IgAN patients and 20 healthy volunteers for renal MRI examinations.  T2* values on BOLD images and ADC, Dt, Dp, fp values on IVIM-DWI images were measured.

Results: Cortical Dt generated the largest area under the curve for differentiating 5-year risk scores ≤ 10% from 5-year risk scores > 10%, and for differentiating T0 from T1/T2.

Conclusions IVIM-DWI and BOLD-MRI can assess the renal function, Oxford classification, and prognostic risk stratification of IgAN patients.

Xia Wang¹, Min Jia¹, Xiuzheng Yue², Queenie Chan², Ruirui Ma¹, Wei Zhang¹, Shanshan Wang¹, Gang Tian¹, Nan He¹, Jiang guang He³, and Yuedong Han^1
1Department of Radiology, Xi'an GaoXin Hospital, Xi'an, China, ²Philips Healthcare, Beijing, China, ³Department of Urology Surgery, Xi'an GaoXin Hospital,, Xi'an, China

Keywords: Kidney, CEST & MT

APTw imaging focuses on cranial tumor studies with significant clinical value, with fewer studies in mid-upper abdomen, because of the movement of organs and B0 fields inhomogeneous. Previously used showed that intermittent breath-hold (IBH) could to improve the image quality of 3D-APTw imaging in healthy adults. In this study, we used the IBH respiratory compensation mode to initially explore the feasibility of 3D-APTw imaging of renal tumors. The findings suggested that 3D-APTw imaging with IBH can be not only used for imaging renal tumors, but also showed potential values in renal tumors diagnosis and differential diagnosis.

View Presentation Video

Sang Youn Kim¹ and Jeong Yeon Cho^1
1Radiology, Seoul National University Hospital, Seoul, Korea, Republic of

Keywords: Kidney, Quantitative Imaging

Quantitative MRI sequences could be a noninvasive assessment modality in the diagnosis and evaluation of CKD. In particular, T1ρ may be a suitable MR sequence parameter to assess renal parenchymal fibrosis in a quantitative manner. Moreover, monitoring the change in common metabolites using MRS may reflect the alteration of osmolality in the renal medulla in CKD.

View Presentation Video

Luis Carlos Sanmiguel Serpa¹, Amber Hillaert², and Pim Pullens^3
1Radiology, Ghent University, AALTER, Belgium, ²Morphology, Imaging, Orthopedics, Rehabilitation and Nutrition, Ghent University, Ghent, Belgium, ³Ghent Institute of Functional and Metabolic Imaging, Ghent University, Ghent, Belgium

Keywords: Kidney, Arterial spin labelling

Arterial Spin Labeling (ASL) is a noninvasive fMRI technique that may evaluate canine renal condition. To date, however, the parameters to estimate the renal blood flow (RBF) have not been calculated on dogs. Therefore, λ: blood-tissue water partition coefficient in kidneys and canine  at 3T were measured. Measured λ was 0.91[mL/g] and  was 1463 ms. With these new values RBF maps were calculated and then compared against the RBF maps given by the scanner. This pilot study demonstrates the relevance of the ASL parameters estimation for the RBF maps calculation in canine kidney.

LING ZHANG ¹, Sheng Xie¹, Xiuzheng Yue², Zhenshan Ding¹, Xiaoqi Pu¹, and Yiwei Li^1
1China-Japan friendship hospital, Beijing, China, ²Philips Healthcare, Beijing, China

Keywords: Kidney, Transplantation

Kidney Transplanted is the therapyfor patients with end-stage renal disease. Arterial spin labeling (ASL) uses magnetic labeling of water from inflowing blood as an endogenous tracer to acquire maps of absolute regional perfusion. The purpose of this study was to evaluate the blood flow of transplanted kidney allograft using a 1.5T MRI with ASL technique and determine whether renal allograft cortical perfusion by ASL could be a non-invasive biomarker for identifying kidney allograft function. Our data showed DWI and 3D-pCASL could be quantitative non-invasive tools for the clinical translation of identifying kidney allografts with subclinical pathology in further study.