(Artificial) Intelligence in the Body
Body Wednesday, 19 May 2021

Oral Session - (Artificial) Intelligence in the Body
Body
Wednesday, 19 May 2021 14:00 - 16:00
  • Deep learning based radial de-streaking for free breathing time resolved volumetric DCE MRI
    Sagar Mandava1, Xinzeng Wang2, Ty Cashen3, Tetsuya Wakayama4, and Ersin Bayram2
    1GE Healthcare, Atlanta, GA, United States, 2GE Healthcare, Houston, TX, United States, 3GE Healthcare, Madison, WI, United States, 4GE Healthcare, Hino, Japan
    Radial imaging is becoming increasingly popular but is plagued by streak artifacts that often arise from undersampling which can lead to poor image quality. We demonstrate a combination of the spatial and temporal DL processing that enables high quality high spatio-temporal imaging.
    Figure 2: Demonstrating the impact of DL processing. Four phases of a DCE scan are shown with arrows highlighting areas where DL delivers improved performance. Best viewed on a high brightness setting on the monitor
    Figure 1: A) Offline trained spatial DL processing in which streaks are suppressed in the spatial dimension, B) Online trained temporal DL processing where a network is trained on the fly for suppressing streaks in the time domain, C) Flow chart of the overall workflow of the proposed method
  • Motion Analysis in Fetal MRI using Deep Pose Estimator
    Junshen Xu1, Esra Abaci Turk2, Borjan Gagoski2, Polina Golland3, P. Ellen Grant2,4, and Elfar Adalsteinsson5
    1Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, MA, United States, 2Fetal-Neonatal Neuroimaging and Developmental Science Center, Boston Children’s Hospital, Boston, MA, United States, 3Computer Science and Artificial Intelligence Laboratory, Massachusetts Institute of Technology, Cambridge, MA, United States, 4Harvard Medical School, Boston, MA, United States, 5Massachusetts Institute of Technology, Cambridge, MA, United States
    We proposed a framework for fetal motion analysis using a deep fetal pose estimator and used it to study the effects of gestational age and maternal position on fetal motion.
    Figure 3. The examples of extracted fetal pose of two different subjects in the dataset and the plots of the knee and elbow angles and the corresponding angular velocities.
    Figure 4. Scatter plots of the three motion metrics vs. gestational age.
  • Automatic segmentation of uterine endometrial cancer on MRI with convolutional neural network
    Yasuhisa Kurata1, Mizuho Nishio1, Yusaku Moribata2, Aki Kido1, Yuki Himoto1, Koji Fujimoto3, Masahiro Yakami2, Sachiko Minamiguchi4, Masaki Mandai5, and Yuji Nakamoto1
    1Diagnostic Imaging and Nuclear Medicine, Kyoto university hospital, Kyoto, Japan, 2Preemptive Medicine and Lifestyle-Related Disease Research Center, Kyoto university hospital, Kyoto, Japan, 3Real World Data Research and Development, Graduate School of Medicine Kyoto University, Kyoto, Japan, 4Diagnostic Pathology, Kyoto university hospital, Kyoto, Japan, 5Gynecology and Obstetrics, Kyoto university hospital, Kyoto, Japan
    The model developed in this study has achieved high-accuracy automatic segmentation of endometrial cancer on MRI using a convolutional neural network for the first time. Using multi-sequence MR images were important for high accuracy segmentation.

    Figure 5: A representative case of the automatic segmentation

    a: T2-weighted image

    b: diffusion-weighted image (b=1000 s/mm2)

    c: apparent diffusion coefficient map

    d: A result of automatic segmentation of endometrial cancer overlaid on T2-weighted image

    The tumor was well segmented despite the presence of hematometra (a:*) (Dice similarity coefficient=0.808).

    Figure 3: The segmentation accuracy of our model for the test datasets with each MRI sequences as input data

    Data are presented with mean±standard deviation

    T2WI: T2-weighted image

    DWI: diffusion-weighted image

    ADC: apparent diffusion coefficient

    Multi: T2WI, DWI, and ADC map

    DSC: Dice similarity coefficient

    PPV: positive predictive value

    NPV: negative predictive value

  • Evaluation of Data Augmentation Methods for Autonomous Segmentation of Placental Volume for Detecting Viral Complications
    Thomas Lilieholm1, Ruiming Chen1, Ruvini Navaratna1, Daniel Seiter1, Walter F Block1,2,3, and Oliver Wieben1,2,3
    1Medical Physics, University of Wisconsin at Madison, Madison, WI, United States, 2Biomedical Engineering, University of Wisconsin at Madison, Madison, WI, United States, 3Radiology, University of Wisconsin at Madison, Madison, WI, United States
    ML-driven autonomous segmentation of placenta can be used in quantification of Zika virus infection biomarkers. Due to scarcity, we augmented data via geometric transform, finding rotation and reflection to yield improved segmentation models.
    The segmentation produced for a single representative slice from the validation dataset, showing the results from training without any augmentation (left) versus training with augmentation via rotation. Comparison shows both a reduction in false positive, or improperly segmented pixels, and an increase in true positive, or properly-segmented pixels.
    The geometric transformations applied to the original dataset to produce augmented data. Each original scan was subject to each type of transform twice, with different parameters. Rotation was up to ±30° about the physical z-axis. Translation was up to ±30 pixels along the physical x- or y-axes. Reflection was across either the physical x- or y-axes. Rescaling brought the placenta up to 500x500 pixels, or down to 150x150 pixels from the original 256x256 scan. Operations were performed using Matlab6.
  • Automated Image Prescription for Liver MRI using Deep Learning
    Ruiqi Geng1,2, Mahalakshmi Sundaresan3, Jitka Starekova1, Collin J Buelo1,2, Nikolaos Panagiotopoulos1, Marcin Ignaciuk1, Thekla Helene Oechtering1, Scott B Reeder1,2,4,5,6, and Diego Hernando1,2
    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, 4Medicine, University of Wisconsin-Madison, Madison, WI, United States, 5Emergency Medicine, University of Wisconsin-Madison, Madison, WI, United States, 6Biomedical Engineering, University of Wisconsin-Madison, Madison, WI, United States
    AI-based automated prescription for liver MRI was demonstrated in a wide variety of clinical datasets, with the potential to advance the development of fully free-breathing, single button-push MRI of the liver. 
    Figure 5. Accuracy of 3D liver localization and axial prescription, including the 6 edges: right (R), left (L), posterior (P), anterior (A), inferior (I), superior (S). Positive (negative) values, ie: purple (green) shading, indicate missing (excess) volume in the automated localization. Whole-liver coronal and sagittal prescription accuracy is similar to axial (not shown due to the additional considerations for coronal/sagittal S/I coverage). S/I coverage shows better accuracy than R/L coverage, possibly due to lower signal in the left lobe and complex anatomy.
    Figure 3. Detection of the liver in each localizer orientation across several patients. a) In most cases, the liver volume was covered accurately by automated prescription. b) Missed coverage in the tip of the lateral segment of the left lobe due to insufficient axial localizer slices. The automated prescription aligned well with the manual annotation in patients with focal lesions (c), ascites (d), and/or cirrhosis (e). f) In patients with splenomegaly, the spleen abuts the liver and its signal level is close to that of liver, but the proposed method was able to select only the liver.
  • Deep Learning-based Adaptive Image Combination for Signal-Dropout Suppression in Liver DWI
    Fasil Gadjimuradov1,2, Thomas Benkert2, Marcel Dominik Nickel2, and Andreas Maier1
    1Pattern Recognition Lab, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany, 2Magnetic Resonance, Siemens Healthcare GmbH, Erlangen, Germany
    Signal-dropouts caused by pulsation can affect a large portion of repetitions in liver DWI, compromising its diagnostic value. Rather than computing a uniform average, we propose to locally suppress signal-dropouts by an adaptive average using weight maps estimated by a CNN.
    Figure 3: First two columns: eight repetitions of a DW liver slice (b = 800 s/mm2) with manually assigned labels. Last two columns: corresponding weight maps produced by the network after processing all patches using a sliding window. Locations with lower weight approximately coincide with image regions affected by signal-dropouts.
    Figure 4: Qualitative analysis of uniform and CNN-based adaptive averaging, compared to the reference image obtained from averaging clean repetitions only. Using the repetitions and weight maps from Figure 3, the proposed method is able to recover signal which leads to higher agreement with the reference as confirmed by the difference maps (5x) as well as the normalized root-mean-squared error (NRMSE) and structural similarity (SSIM). Accordingly, elevated ADC values in the affected liver lobe (ROI, yellow box) are corrected with the proposed method.
  • Utility of Texture Analysis on Quantitative Susceptibility Maps to Stage Hepatic Fibrosis
    FengLing Gan1, Shuohui Yang2, Feng Xing3, Zheng Qu1, Gaiying Li1, Chenyao Yang2, Rongfang Guo2, Jiling Huang2, Fang Lu2, Caixia Fu4, Xu Yan4, Kelly Gillen5, Yi Wang5, Chenghai Liu3, Songhua Zhan2, and Jianqi Li1
    1Shanghai Key Laboratory of Magnetic Resonance, School of Physics and Electronic Science, East China Normal University, shanghai, China, 2Department of Radiology, Shuguang Hospital Affiliated to Shanghai University of Traditional Chinese Medicine, shanghai, China, 3Department of Liver Diseases, Shuguang Hospital Affiliated to Shanghai University of Traditional Chinese Medicine, shanghai, China, 4MR Collaboration NE Asia, Siemens Healthcare, shanghai, China, 5Department of Radiology, Weill Medical College of Cornell University, New York, NY, United States
    This study measured the texture features on susceptibility maps in patients with chronic liver diseases.There were significant differences between two cohorts in DifVarnc , entropy , contrast , The DifVarnc of susceptibility maps yielded the optimal performance  (AUC = 0.848, P=0.002)

    Table 1. The results of the ROC analyses of the texture features extracted from QSM between cohorts of significant and non-significant hepatic fibrosis patients.

    Only the results with statistically significant differences between the cohorts of significant and non-significant hepatic fibrosis are shown. AUC = area under receiver operating characteristic curve, SS = sensitivity, SP = specificity. DifVarnc = difference of variance, DifEntrp = difference of entropy, AngScMom = angular second moment, InvDfMom = inverse different moment.

    Fig 1. The second-order texture analysis of susceptibility maps between cohorts of significant and non-significant hepatic fibrosis. The vertical axes are dimensionless quantities which are calculated from the gray level co-occurrence matrix. * indicates p < 0.05; ** indicates p < 0.01. DifVarnc = difference of variance, DifEntrp = difference of entropy, AngScMom = angular second moment, InvDfMom = inverse different moment.
  • Discriminative feature learning and adaptive fusion for the grading of hepatocelluar carcinoma with Contrast-enhanced MR
    Wu Zhou1, Shangxuan Li1, Wanwei Jian1, Guangyi Wang2, Lijuan Zhang3, and Honglai Zhang1
    1School of Medical Information Engineering, Guangzhou University of Chinese Medicine, Guangzhou, China, 2Department of Radiology, Guangdong General Hospital, Guangzhou, China, 3Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China
    The purpose is to address the problem of multimodal fusion with contrast-enhanced MR for grading hepatocellular carcinoma(HCC).  We proposed a discriminative feature learning and adaptive fusion method in the framework of deep learning architecture for improving the fusion performance. 
    Figure 2. The framework of the proposed multimodal fusion method.
    Figure 1. Contrast-enhanced MR images of a 59-year-old man with pathologically confirmed low-grade HCCs (grade II, white arrow) (a) the pre-contrast phase (b) the arterial phase (c) the portal vein phase (d) the delayed phase.
  • Automatic Detection of Small Hepatocellular Carcinoma (≤2 cm) in Cirrhotic Liver based on Pattern Matching and Deep Learning
    Rencheng Zheng1, Luna Wang2, Chengyan Wang3, Xuchen Yu1, Weibo Chen4, Yan Li5, Weixia Li5, Fuhua Yan5, He Wang1,3, and Ruokun Li5
    1Institute of Science and Technology for Brain-Inspired Intelligence, Fudan University, Shanghai, China, 2Department of Radiology, Shanghai Chest Hospital, Shanghai, China, 3Human Phenome Institute, Fudan University, Shanghai, China, 4Market Solutions Center, Philips Healthcare, Shanghai, China, 5Department of Radiology, Ruijin Hospital, Shanghai, China
    Feasibility of automatic detection and segmentation of small hepatocellular carcinoma (≤2 cm) in cirrhotic liver based on diffusion-weighted imaging and dynamic contrast-enhanced images with pattern matching and deep learning model.
    Figure 1. Overall framework of the proposed PM-DL model, including a) image registration and liver segmentation, b) screening of suspicious lesions in DWI images, and c) identification and segmentation of true lesions in DCE images.
    Figure 4. Two representative sHCCs correctly detected by the PM-DL model in the external test cohort, a) LR-4 according to LI-RADS v2018, which can be regarded as HCCs, Edmondson–Steiner grade III; b) LR-M according to LI-RADS v2018, which cannot be regarded as HCCs, Edmondson–Steiner grade II. (Yellow box: extracted image patch, green contour: manually labeled mask, yellow contour: model predicted mask).
  • Peritumoral Dilation Radiomics of Gd-EOB-DTPA MRI Predicts Early Relapse in Hepatocellular Carcinoma Without Macrovascular Invasion
    Huan-Huan Chong1,2, Yu-Da Gong3, Lei Chen4, Xian-Pan Pan4, Ai-E Liu4, Chun Yang2, and Meng-Su Zeng1,2,5
    1Shanghai Institute of Medical Imaging, Shanghai, China, 2Department of Radiology, Zhongshan Hospital, Fudan University, Shanghai, China, 3Department of General Surgery, Zhongshan Hospital, Fudan University, Shanghai, China, 4Shanghai United Imaging Intelligence Co., Ltd., Shanghai, China, 5Department of Medical Imaging, Shanghai Medical College, Fudan University, Shanghai, China
    The peritumoral dilation radiomics of Gd-EOB-DTPA MRI is an excellently preoperative biomarker for stratifying the HCC patients at high risk of early recrudescence after hepatectomy. 
    Flowchart of study population.
    Table 1. The discrimination of single sequences in predicting 2-year recurrence with logistic regression.
Back to Top
Digital Poster Session - Even More (Artificial) Intelligence in the Body
Body
Wednesday, 19 May 2021 15:00 - 16:00
  • FAST 3D vs. Compressed Sensing vs. Parallel Imaging: Image Quality Improvement on MRCP with and without Deep Learning Reconstruction
    Takahiro Matsuyama1, Yoshiharu Ohno1,2, Kaori Yamamoto3, Kazuhiro Murayama2, Masato Ikedo3, Masao Yui3, Akiyoshi Iwase4, Takashi Fukuba4, Satomu Hanamatsu1, Yuki Obama1, Takahiro Ueda1, Hirotaka Ikeda1, and Hiroshi Toyama1
    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
    DLR method (AiCE) can significantly improve image quality of MRCP at all protocols.  Image quality of FAST 3D is superior to that of Compressed SPEEDER and considered as compatible with conventional SPEEDR on MRCP.

    Figure 1. 71-year old patient with intraductal papillary mucinous neoplasm. (First line from L to R: MRCPFAST 3D with AiCE, MRCPCompressed SPEEDER with AiCE and MRCPconventional SPEEDER with AiCE; second line from L to R: MRCPFAST 3D without AiCE, MRCPCompressed SPEEDER without AiCE and MRCPconventional SPEEDER without AiCE).

    MRCPFAST 3D and MRCPconventional SPEEDER more clearly depict intrahepatic bile duct as well as main pancreatic duct than MRCPCompressed SPEEDER, when applied AiCE or not. In addition, AiCE was able to improve image quality of MRCP obtained by each technique.

    Figure 4. Compared results of interobserver agreement and qualitative image qualities among all methods.

    Interobserver agreement on each method was ranged between 0.44 and 0.77. When applied AiCE, each index on MRCP obtained by each technique and reconstructed with AiCE were significantly higher than those without AiCE (p<0.05). On each index, MRCPCompressed SPEEDER with and without AiCE were significantly lower than MRCPFAST 3D or MRCPconventional SPEEDER with and without AiCE (p<0.05).

  • The value of radiomics-based model on contrast-enhanced MRI for predicting microvascular invasion in HCC before Partial Hepatectomy
    Tao Lin1, Ailian Liu1, Lihua Chen1, Qingwei Song1, Renwang Pu1, Ying Zhao1, Xue Ren1, and yan guo2
    1Department of Radiology, the First Affiliated Hospital of Dalian Medical University, Dalian, China, 2GE Healthcare, Beijing, China
    Based on enhanced MRI images, we established an radiomics model to predict microvascular invasion of hepatocellular carcinoma, contributing to precise decisions regarding treatment.
    Figure1. Receiver operating characteristic curves (ROC) of the training (A) and validation (B) cohort. AUC, area under the receiver operating characteristic curve.
    Table 1. Radiomics features extracted based on enhanced MRI
  • Deep learning-based detection of liver disease using MRI
    Mark A Pinnock1,2, Yipeng Hu1,2, Alan Bainbridge3, David Atkinson4, Rajeshwar P Mookerjee5, Stuart A Taylor4, Dean C Barratt1,2, and Manil D Chouhan4
    1Centre for Medical Image Computing, University College London, London, United Kingdom, 2Wellcome/EPSRC Centre for Interventional and Surgical Sciences, University College London, London, United Kingdom, 3Department of Medical Physics and Biomedical Engineering, University College London Hospitals NHS Foundation Trust, London, United Kingdom, 4Centre for Medical Imaging, Division of Medicine, University College London, London, United Kingdom, 5Institute for Liver and Digestive Health, Division of Medicine, University College London, London, United Kingdom
    Deep learning-based detection of liver disease from T2-weighted MRI sequences/relevant organ segmentation mask only is feasible.
    Figure 1: GradCAM visualisations for two false positive classifications
    Figure 2: GradCAM visualisations for two true positive classifications
  • A Two-Stage Deep Learning Model for Accurate Vessel Segmentation and Reconstruction in the MRI of Live
    Xu Luo1,2, Ailian Liu3, Yu Yao1,2, Ying Zhao3, Zhebin Chen1,2, Meng Dou1,2, and Han Wen1,2
    1Chengdu Institute of Computer Application, Chinese Academy of Sciences, Chengdu, China, 2University of Chinese Academy of Sciences, Beijing, China, 3Department of Radiology, the First Affiliated Hospital of Dalian Medical University, Dalian, China
    We use a two-stage segmentation method to complete the segmentation of intrahepatic portal vein, and complete the volume calculation and three-dimensional reconstruction of intrahepatic portal vein
    Figure 3: Visualization of segmentation results
    Figure 1: Overall framework
  • Deep Learning 3D Convolutional Neural Network for Noninvasive Evaluation of Pathologic Grade of HCC Using Contrast-enhanced MRI
    Ying Zhao1, Han Wen2,3, Ailian Liu1, Yu Yao2,3, Tao Lin1, Qingwei Song1, Xin Li4, Yan Guo4, and Tingfan Wu4
    1The First Affiliated Hospital of Dalian Medical University, Dalian, China, 2Chengdu Institute of CoChinese Academy of Sciences, Chengdu, China, 3University of Chinese Academy of Sciences, Beijing, China, 4GE Healthcare (China), Shanghai, China
    3D-CNN based on CE-MR images was demonstrated to be capable to evaluate pathologic grade of HCC treated with surgical resection, which will provide more prognostic information and facilitate clinical management.
    Table 1. The architecture of ResNet-50
    Figure 2. The feature extraction process of ResNet-50.
  • T2WI liver MRI with deep learning-based reconstruction: a clinical feasibility study in comparison to conventional T2WI liver MRI
    Ruofan Sheng1, Liyun Zheng2, Shu Liao3, Yongming Dai2, and Mengsu Zeng1
    1Department of Radiology, Zhongshan Hospital, Shanghai, China, 2United Imaging Healthcare, Shanghai, China, 3Shanghai United Imaging Intelligence, Shanghai, China
    Compared with the conventional T2WI, the T2WI sequence with deep learning-based reconstruction showed promising performance as it provided significantly better image quality and lesion detectability within a relatively shorter acquisition time. 
    Figure 1. Network structure of the deep learning-based fast MR reconstruction framework used in this study.
    Figure 4. Example of lesion analysis.
  • Model-based Deep Learning Reconstruction using Folded Image Training Strategy (FITS-MoDL) for Liver MRI Reconstruction
    Satoshi Funayama1,2, Utaroh Motosugi3, and Hiroshi Onishi1
    1Department of Radiology, University of Yamanashi, Yamanashi, Japan, 2Graduate School of Medicine, University of Yamanashi, Yamanashi, Japan, 3Department of Radiology, Kofu-Kyoritsu Hospital, Yamanashi, Japan
      Model-based Deep Learning Reconstruction using Folded Image Training Strategy (FITS-MoDL) improved image quality and memory consumption during network training in liver MRI.
    Flow diagram of network training with folded image training strategy (FITS) and image reconstruction. In training with FITS, images for training were folded by factor of 2 to reduce memory consumption.
    A representative case. (CS) total variation regularized compressed sensing showed some aliasing on liver parenchyma. A few aliasing is remained in the conventional model-based deep learning reconstruction (conv-MoDL), while it is removed in the model-based deep learning reconstruction using FITS (FITS-MoDL).
  • The value of Radiomics combined with Machine Learning in the staging of liver Fibrosis
    Fengxian Fan1, Weiting Huang2, Yanli Jiang1, Wanjun Hu1, Jing Zhang1, and Jialiang Ren3
    1LanZhou University Second Hospital, LanZhou, China, 2LanZhou University, LanZhou, China, 3GE Healthcare, Shanghai, China
    In this study, 244 people had liver pathologic and MRI were divided into training and testing cohorts to developed and validated radiomics models for differentiation of low(0-2) from high(3-4) stage fibrosis. The results showed  radiomics models  of T1WI had a powerful ability to stage fibrosis.
    Figure 2. a. ROC curve analysis of radiomics models in the testing cohort (AUC=0.85). b. ROC curve analysis of Fibroscan(AUC=0.83), APRI(AUC=0.68), FIB-4(AUC=0.69).
    Table 1.The eighteen valuable features.
  • Few-shot deep learning for kidney segmentation
    Junyu Guo1 and Ivan Pedrosa1
    1Radiology, UT southwestern medical center, Dallas, TX, United States
    In this study, we investigated the feasibility of kidney segmentation using deep learning models trained with MR images from only a few subjects. We tested the hypothesis that few-shot deep learning may achieve accurate kidney segmentation.
    Figure 2. Comparison between two results (Seg1 vs. Seg2) of Dice coefficients using the different trained models. A.) Dice coefficient plots from a model trained using one subject (1Subj); B.) using three subjects (3Subj); C.) using the six subjects (6Subj). D.) the bar plots for the above three cases. Seg1 indicates the results predicted using the first network alone; Seg2 indicates the results by using the two networks in Fig. 1. Gray areas in A-C indicate the training data sets.
    Figure 1. Diagram of two neural networks including UResNet1 and UResnet2. The inputs for UResNet1 are three consecutive slices; the inputs for UResNet2 are three consecutive slices and one predicted mask for the third slice. Seg1 indicates the output of the first network; Seg2 indicates the output of the whole networks. The dotted and dashed lines indicate the relationship in the prediction stage.
  • Deep learning based kidney segmentation for high temporal resolution tracking renal size changes during sequential gas challenges
    Kaixuan Zhao1,2, Joao dos Santos Periquito3, Thomas Gladytz2, Kathleen Cantow3, Luis Hummel3, Jason Millward2, Sonia Waiczies2, Erdmann Seeliger3, Yanqiu Feng1, and Thoralf Niendorf2,4
    1School of Biomedical Engineering, Southern Medical University, Guangzhou, China, 2Berlin Ultrahigh Field Facility (B.U.F.F.), Max Delbruck Center for Molecular Medicine in the Helmholtz Association, Berlin, Germany, 3Institute of Physiology, Charité - Universitätsmedizin Berlin, Berlin, Germany, 4Experimental and Clinical Research Center, a joint cooperation between the Charité Medical Faculty and the Max Delbrück Center for Molecular Medicine in the Helmholtz Association, Berlin, Germany
    With the progression of hypoxia, renal size decreased by ~10%. During the reoxygenation phase renal size rapidly recovered to baseline. A comparison between renal size and renal T2* demonstrates that rapid renal size recovery is paralleled by T2* recovery.  
    Figure 1. Flow chart of experiment design
    Figure 4. Comparison between renal layer’s T2* and measured renal size changes.
  • Is it Feasible? IVIM-DWI and T2WI-based Texture Analysis Predicting Histological Types of Cervical Carcinoma Before Operation
    Jiang-Ning Dong1 and Bin Shi1
    1The First Affiliated Hospital of USTC, Anhui Provincial Cancer Hospital, Hefei, China

    The combination of IVIM-DWI and T2WI-based texture features had good predictive performance to evaluate different histological types of cervical carcinoma, especially for cervical squamous cell carcinoma and adenocarcinoma.

    Fig. 2 Box-plots of IVIM-DWI Parameters.

    Panels A-D represent ADC, D, D* and f values for G-1 (cervical squamous cell carcinoma), G-2 (cervical adenocarcinoma) and G-3 (cervical small cell carcinoma), respectively (p < 0.05).

    Fig. 3 ROC Curves of Regression Model for G-1 Compared to G-2, and Individual Independent factors (D, D* and Gray Level Variance).
  • Fully Automated Pelvic Bones Segmentation in Multiparameter MRI Using a 3D Convolutional Neural Network
    xiang liu1, chao han1, and xiaoying wang1
    1department of radiology, peking university first hospital, Beijing, China
    The 3D U-Net CNN showed good quantitative and qualitative performances in the segmentation of pelvic bones on mpMRI images, which may provide reliable skeletal geometric information for subsequent detection of pelvic tumor bone metastases
    Figure 3. Examples of the comparison between CNN-predicted and manual segmentation. (a) Section examples of eight bones on DWI image; (b) The corresponding overlapping images between manual segmentation (white background) and CNN-predicted segmentation; (c) Section examples of eight bones on ADC image; (d) The corresponding overlapping images between manual segmentation (white background) and CNN-predicted segmentation .CNN: Convolution neural network.
    Figure 2. The SCORE system and evaluation criteria on DWI and ADC images. Condition A refers to that the location of the predicted CD is consistent with manual CD, and the range of the predicted CD is larger than (A1) or less than (A2) the manual CD, or partially overlaps with the manual CD (A3). CD: connected domain, which is defined as the label with a continuous structure in 3D space. DSC: Dice similarity coefficient.
  • Motion Robust High-Resolution Pelvic Imaging using PROPELLER and Deep Learning Reconstruction
    Ali Pirasteh1, Lloyd Estkowski2, Daniel Litwiller3, Ersin Bayram4, and Xinzeng Wang5
    1Department of Radiology, UW Madison, Madison, WI, United States, 2Global MR Applications & Workflow, GE Healthcare, Madison, WI, United States, 3Global MR Applications & Workflow, GE Healthcare, Denver, CO, United States, 4Global MR Applications & Workflow, GE Healthcare, Houston, TX, United States, 5GE Healthcare, Houston, TX, United States
      PROPELLER T2-weighted images of pelvis with DL reconstruction resulted in improved image quality, including improved SNR, in-plane resolution and robustness to motion.
    Figure 4: FSE and PROPELLER images reconstructed using deep-learning based reconstruction methods. Deep-learning based reconstruction method improved the SNR and in-plane resolution of FSE images, but the motion artifacts were not removed. Due to the robustness to motion and deep-learning based reconstruction method, the PROPELLER images showed better SNR, in-plane resolution and less motion artifacts.
    Figure 5: High resolution rectal FSE and PROPELLER images of a patient. Deep learning reconstruction method improved the SNR and in-plane resolution. Due to the robustness to motion, PROPELLER further improved the image sharpness.
  • Deep Learning Reconstruction for DWI with b Values < 5000s/mm2: Improvement of Image Quality and Diagnostic Performance for Prostatic Cancer
    Takahiro Ueda1, Yoshiharu Ohno1,2, Kaori Yamamoto3, Kazuhiro Murayama2, Masato Ikedo3, Masao Yui3, Akiyoshi Iwase4, Takashi Fukuba4, Satomu Hanamatsu1, Yuki Obama1, Hirotaka Ikeda1, and Hiroshi Toyama1
    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
    DLR method is useful for improving image quality and diagnostic performance of DWI without any adverse effect on ADC assessment using a 3T MR system for patients with prostatic cancer. 

    Figure 1. 51-year old patient with prostatic cancer in the left transitional zone

    b: DLR improves image quality of DWIs for each b value. Each DWI shows the prostatic cancer as high signal intensity in the left transitional zone (arrow).

    Figure 1. 51-year old patient with prostatic cancer in the left transitional zone

    c: All ADC maps show the prostatic cancer as a hypointense signal in the left transitional zone (arrow).

  • T2-weighted Pelvic MR Imaging Using PROPELLER with Deep Learning Reconstruction for Improved Motion Robustness
    Mohammed Saleh1, Sanaz Javadi1, Manoj Mathew2, Jong Bum Son3, Jia Sun4, Ersin Bayram5, Xinzeng Wang5, Jingfei Ma3, Janio Szklaruk1, and Priya Bhosale1
    1Radiology, MD Anderson Cancer Center, Houston, TX, United States, 2Radiology, Stanford University, Stanford, CA, United States, 3Imaging Physics, MD Anderson Cancer Center, Houston, TX, United States, 4Biostatistics, MD Anderson Cancer Center, Houston, TX, United States, 5Global MR Applications and Workflow, GE Healthcare, Houston, TX, United States
    Compared to FSE, PROPELLER showed reduced motion artifacts for T2-weighted imaging of pelvis. Deep Learning-based image reconstruction (DL Recon) further improved image quality with better image SNR, increased image sharpness and reduced artifacts.
    Figure 1. The conventional T2-weighted FSE images and PROPELLER images without and with DL are shown. Because the phase encoding direction of AP was often chosen to avoid aliasing and long scan time, conventional T2-weighted FSE is sensitive to motion artifacts in sagittal imaging (A). PROPELLER is robust to motion and minimized the motion artifacts (B), however, the SNR and in-plane resolution were limited by the clinical scan time. DL reconstruction improved the SNR and in-plane resolution without increasing scan time (C).
    Figure 3. This graph shows the scoring of the best sequence on the 4 images by three radiologists. DL 50 and DL 75 were considered to be better than Non-DL and DL25.
  • Quantifying Efficiency and Variability of Clinical MRI Exams with Advanced Analytics Tools
    Sheena Y Chu1, Scott B Reeder1,2,3,4,5, and John W Garrett1,3
    1Department of Medical Physics, University of Wisconsin-Madison, Madison, WI, United States, 2Department of Medicine, University of Wisconsin-Madison, Madison, WI, United States, 3Department of Radiology, University of Wisconsin-Madison, Madison, WI, United States, 4Department of Biomedical Engineering, University of Wisconsin-Madison, Madison, WI, United States, 5Department of Emergency Medicine, University of Wisconsin-Madison, Madison, WI, United States
    We developed an analytics tool to quantitatively assess the utilization of clinical MRI exams. By tracking overall exam efficiency and variability, we assessed sources of inefficiency and variability that contribute to lengthy exam slots. 
    Table 1: MR enterography protocol before and after intervention
    Table 2: The intervention led to an overall decrease in exam length and standard deviation, with a resulting increase in efficiency.
  • Application value of Radiomics Methods Based on DKI Sequence MK Map for Differentiating squamous Cell carcinoma from cervix Adenocarcinoma
    Shifeng Tian1, Ailian Liu1, Yan Guo2, and Yuan Wei1
    1Department of Radiology, The First Affiliated Hospital of Dalian Medical University, Dalian, China, China, 2GE Healthcare, Dalian City, China, China
    A total of 386 imaging features were extracted, and 7 omics characteristics related to cervical cancer pathological classification were obtained by dimension reduction.
    Figure 1:A 56 year old patient with cervical mucinous adenocarcinoma.1A shows T2WI image and the red arrow points to cervical cancer;1B shows the MK pseudo color map of DKI sequence of corresponding layer; 1C shows the ROI outline of corresponding layer, and the red area shows the coverage area of tumor parenchyma.
  • Radiomics Based on MR Imaging of Rectal Mucinous Adenocarcinoma: Assess Treatment Response to Neoadjuvant Chemoradiotherapy
    Fu Shen1, Minglu Liu1, Zhihui Li1, Xiaolu Ma1, Jianping Lu1, and Yuwei Xia2
    1Changhai Hospital, Shanghai, China, 2Huiying Medical Technology Co., Ltd., Beijing, China
    We developed a radiomics model with excellent performance for individualized, noninvasive prediction of tumor regression in patients with rectal mucinous adenocarcinoma (RMAC).
    The ROC curve for LR classifier
  • Breast MRI radiomic shape features for the prediction of neoadjuvant therapy response
    Wen Li1, Rohan Nadkarni1, David C Newitt1, Bo La Yun1,2, Deep Hathi1, Alex Nguyen1, Natsuko Onishi1, Lisa J Wilmes1, Ella F Jones1, Jessica Gibbs1, Teffany Joy Bareng1, Bonnie N Joe1, Elissa Price1, Rita Mukhtar1, John Kornak1, Efstathios Gennatos1, I-SPY 2 Consortium3, Laura J Esserman1, and Nola M Hylton1
    1University of California, San Francisco, San Francisco, CA, United States, 2Seoul National University Bundang Hospital, Seoul, Korea, Republic of, 3Quantum Leap Healthcare Collaborative, San Francisco, CA, United States
    In the prediction of pathologic complete response of breast cancer in neoadjuvant setting, addition of radiomic shape features to FTV on MRI showed improvements in the performance over FTV alone, especially at the early treatment time point.
    Fig. 4 AUCs for predicting pCR using logistic regression model at T0 (top) and T1 (bottom). The number of patients and pCR rate included in analysis at T0 were: 941 (35%), 368 (18%), 149 (39%), 81 (68%), 343 (43%) for full, HR+/HER2-, HR+/HER2+, HR-/HER2+, HR-/HER2- respectively. The numbers at T1 were: 904 (34%), 357 (18%), 142 (39%), 74 (66%), 331 (42%). FTV – functional tumor volume, HR – hormone receptor, HER2 – human epidermal growth factor receptor 2.
    Fig. 3 Correlation map among FTV and shape features at T0.
  • The nomogram of MRI-based radiomics with complementary visual features by machine learning improves stratification of glioblastoma patients
    ZHENYU SHU1, YUYUN XU1, and YONG ZHANG2
    1Zhejiang Provincial People’s Hospital, Hangzhou, China, 2MR Research, GE healthcare (China), SHANG HAI, China
    The nomogram had a survival prediction accuracy of 0.878 and 0.875, a specificity of 0.875 and 0.583, and a sensitivity of 0.704 and 0.833, respectively, in the training and test set
    (A)ROC curves for nomogram, radiomic signature, age and meninges predicting OS in validation set. (B) The stratification performance of the nomogram in validation set.
    (A) and (B) show the ROC curves of the nomograms in the training and test sets; (C) and (D) show the calibration curves of the nomograms in the training and test sets; and (E) and (F) show the DCA curve of the nomograms in the training and test sets.