Tianwen Xie¹, Tingting Jiang¹, Caixia Fu², Marcel Dominik Nickel³, Weijun Peng¹, and Yajia Gu^1
1Radiology, Fudan University Shanghai Cancer Center, Shanghai, China, ²MR Applications Development, Siemens Shenzhen Magnetic Resonance Ltd., Shenzhen, China, ³MR Application Predevelopment, Siemens Healthcare GmbH, Erlangen, Germany

In this study, we analyzed the relationship between CAIPIRINHA-Dixon-TWIST-VIBE (CDTV)-derived DCE parameters and clinicopathological factors, including molecular subtypes of invasive ductal carcinoma (IDC). Our results showed that the values of semi-quantitative parameters were similar for various prognostic factors and molecular subtypes of IDC. However, the values of quantitative inline pharmacokinetic kinetic parameters, k_ep_minimum and k_ep_difference, were significantly different for the luminal A and HER2-positive subtypes. AUC value for accurately distinguishing luminal A and HER2-positive subtypes based on combined k_ep_minimum and k_ep_difference values was 0.820. Therefore, our study demonstrates that CDTV-derived quantitative parameters are associated with molecular subtypes of IDC.

Chunping Mao¹, Mengzhu Wang², Xu Yan², Guang Yang³, Xiang Zhang¹, and Jun Shen^1
1Department of Radiology, Sun Yat-Sen Memorial Hospital, Sun Yat-Sen University, Guangzhou, China, ²MR Scientific Marketing, Siemens Healthineers, Guangzhou, China, ³MR Scientific Marketing, School of Physics and Electronic Science, Shanghai, China

This study explored the feasibility of applying multiple diffusion models based on the same diffusion spectrum imaging (DSI) data into preoperative prediction of HER-2 status in patients with breast cancer (BC). The results showed that some diffusion parameters can discriminate HER-2 positive BC patients, but apparent diffusion coefficient (ADC) based on conventional mono-exponential model can not. This study suggests that the simultaneously application of multiple diffusion models from DSI data is feasible  for diagnosis of patients with BC, which has certain clinical value of preoperative prediction of HER-2 status and is worthy of further research.

Mana Kato¹, Kazuo Kodaira¹, Yasuhiro Goto¹, Yasutomo Katsumata², Mai Nishihara², Masami Yoneyama², Isao Shiina¹, Yutaka Hamatani¹, Takumi Ogawa¹, Michinobu Nagao³, and Shuji Sakai^3
1Radiological Services, Tokyo Woman's Medical University Hospital, Tokyo, Japan, ²Philips Japan, Tokyo, Japan, ³Diagnostic imaging & Nuclear Medicine, Tokyo Woman's Medical University Hospital, Tokyo, Japan

DWI provides beneficial information that has high diagnostic potential of breast lesions. DWIBS can suppresses the background signals including fat tissues effectively, but there are often insufficient fat suppression cases because of non-adaptation TInull were used. In this study we demonstrated that the actual TI for nulling breast fat in DWI is patient-specific and dynamic TI scout scan (DynTI) is useful to explore patient-adaptive TI that can provide more beneficial to increase the robustness of breast DWIBS.

Jing Zhang¹, Chenao Zhan², Xu Yan³, Yang Song³, Yihao Guo⁴, Tao Ai², and Guang Yang^1
1East China Normal University, Shanghai key lab of magnetic resonance, shanghai, China, ²Tongji Medical College, Huazhong University of Science and Technology, Department of Radiology, Tongji Hospital, Wuhan, Hubei Province, China, ³Siemens Healthcare, MR Scientific Marketing, shanghai, China, ⁴Siemens Healthcare, MR Collaboration, Guangzhou, China

         Multi-parametric MRI (mp-MRI) radiomics can distinguish breast mass effectively, but requires breast lesion segmentation first, which is subjective and laborious for radiologists. To overcome this problem, we combined nnUnet and radiomics analysis as an automatic model for breast lesion classification. In the test cohort, the breast lesion segmentation model achieved mean dice of 0.835, and the classification model achieved an AUC of 0.891. We found that the nnU-Net can delineatey lesions accurately based on dynamic contrast-enhanced (DCE, TWIST-VIBEs)), and mp-MRI radiomics features extracted from the auto-segmented lesions can be  used to classfy breast lesions accurately.      

Pan Wang¹, Min Li¹, Xiangnan Li¹, Jiyang Zhang¹, Xin Zheng², Chen Zhang³, Stemmer Alto⁴, and Tao Jiang^1
1radiology department, Beijing Chao-Yang Hospital, Beijing, China, ²Urinary Surgery, Beijing Chao-Yang Hospital, Beijing, China, ³MR Scientific Marketing, Siemens Healthcare, Beijing, China, ⁴] MR Application Development, Siemens Healthcare, Erlangen, Germany

Kidney transplantation is currently considered as the preferred treatment for end-stage renal disease, this study aims to explore the performance of BOLD-fMRI and IVIM of renal transplants kidney patients with CKD. The results showed the D value in renal cortical could well identify and distinguish normal kidney and diseased kidney. The R2* value could indirectly reflect deoxyhemoglobin to assess the degree of renal function impairment. BOLD-fMRI and IVIM can provide a more accurate and comprehensive understanding of the functional information of the chronic kidney metabolism.

Michinobu Nagao¹, Taro Ando¹, Yasuhiro Goto¹, Isao Shiina¹, Kazuo Kodaira¹, Masami Yoneyama², Kenichiro Miura¹, Motoshi Hattori¹, and Shuji Sakai^1
1Tokyo Women's Medical University, Tokyo, Japan, ²Philips Japan, Tokyo, Japan

For pediatric patients with end-stage renal failure, kidney transplantation is the only treatment that can provide growth and long-term survival. Although rejection is the common cause of renal dysfunction, there are other causes of chronic transplant nephropathy, in which the transplanted kidney develops fibrosis. Hypoxia is considered as possible causes of this order. We used O2-inhalation T2*-BOLD and dual-VENC 4D flow MRI to assess hypoxia in pediatric transplanted kidneys. In patients with chronic transplant nephropathy, hypoxia was found in the renal medulla and small vessel’s flow was reduced. This analysis is an effective means of monitoring hypoxia in transplanted kidneys.

lihua chen¹, jianmin cui², jinxia zhu³, and wen shen^2
1Tianjin First Center Hospital, tianjin, China, ²Tianjin First Center Hospital, Tianjin, China, ³MR collaboration, Siemens Healthcare Ltd., Beijing, China

Early evaluation and treatment of delayed graft function due to renal cold ischemia reperfusion injury has important clinical significance for prolonging donor kidney preservation times and improving the survival rates of allografts. Arterial spin labeling (ASL) and diffusion-weighted imaging (DWI) have been accepted for microstructure change evaluations of renal function. In this study, transplant recipients were collected for fMRI examination at 14d, 30d and 90d after transplantation. The parameters were compared to evaluate renal microstructure changes among different time points. The results showed ASL and DWI can reflect longitudinal changes of diffusion and perfusion properties for allografts after kidney transplantation . 
 

Xiaojun Yang¹, Xiaohui Duan¹, Mengzhu Wang², Xu Yan², Lingjie Yang¹, Weike Zeng¹, Wei Jiang¹, Yaojun Hu¹, Guang Yang³, and Jun Shen^1
1Department of Radiology, Sun Yat-Sen Memorial Hospital, Sun Yat-Sen University, Guangzhou, China, ²2MR Scientific Marketing, Siemens Healthineers, Guangzhou, China, ³Shanghai Key Laboratory of Magnetic Resonance, School of Physics and Electronic Science, East China Normal University, Shanghai, China

This study evaluated whether continuous-time random-walk (CTRW) diffusion model could be used to distinguish benign from malignant renal tumors, as well as compared the diagnostic effectiveness between the diffusion parameters of CTRW and apparent dispersion coefficient (ADC). The results demonstrated that CTRW diffusion model as a new noninvasive MR technique, was able to differentiate benign from malignant renal masses in vivo, and further improve the diagnostic efficiency of renal malignant masses compared with conventional diffusion imaging.

Yusaku Moribata¹, Yasuhisa Kurata¹, Mizuho Nishio¹, Aki Kido¹, Satoshi Otani¹, Yuki Himoto¹, Naoko Nishio², Akihiro Furuta², Kimihiko Masui³, Takashi Kobayashi³, and Yuji Nakamoto^1
1Department of Diagnostic Imaging and Nuclear Medicine, Kyoto University Graduate School of Medicine, Kyoto, Japan, ²Department of Radiology, Japanese Red Cross Osaka Hospital, Osaka, Japan, ³Department of Urology, Kyoto University Graduate School of Medicine, Kyoto, Japan

This multi-center retrospective study performed automatic segmentation of bladder cancer (BC) on MRI with a convolutional neural network and evaluated the reproducibility of radiomics features. Of the total 170 patients, 140 were used to train our U-net model and 30 were used to evaluate the segmentation performance of the model. Our U-net model achieved a median Dice similarity coefficient of 0.811 in the test dataset and most of the automatically extracted radiomics features showed high reproducibility (median intraclass correlation coefficient: 0.83-0.86). Our model would lead to efficient medical image analysis of BC using the radiomics approach.

Miao Liu¹, Qi Wang¹, Gaofeng Shi¹, Li Yang¹, and Qinglei Shi^2
1The Fourth Hospital of Hebei Medical University, Shi Jiazhuang, China, ²MR Scientific Marketing , Siemens Healthineers Ltd., Beijing, China

To overcome the small data set problems in clinical situations, this paper proposed a pairwise auto encoder (AE) model, which can learn more relationship information among samples to enhance the generalization ability, in predicting the prognosis of concurrent chemo-radiotherapy for LACC and demonstrated potential in this field. 

Wan Dong¹, Shifeng Tian¹, Lihua Chen², and Ailian Liu^2
1Radiology, the First Affiliated Hospital of Dalian Medical University, Dalian, China, ²the First Affiliated Hospital of Dalian Medical University, Dalian, China

Immunohistochemistry is considered the most common method to determine the MSI status of EC, but this method is usually performed after surgery. Here, diffusion kurtosis imaging (DKI) with multiple quantitative parameters was used to predict microsatellite instability (MSI) status of endometrial carcinoma (EC). Results showed that there were good efficiency of DKI to distinguish MSI status of EC (The AUC of MK, Ka, MD, and Dr values was 0.763,0.729,0.731, 0.748, respectively).

Takahiro Ueda¹, Yoshiharu Ohno¹, Kaori Yamamoto², Natsuka Yazawa², Ikki Tozawa³, Masayuki Sato³, Motohiro Katagiri³, Masato Ikedo², Masao Yui², Hiroyuki Nagata¹, Kazuhiro Murayama⁴, and Hiroshi Toyama^1
1Radiology, Fujita Health University School of Medicine, Toyoake, Japan, ²Canon Medical Systems Corporation, Otawara, Japan, ³Radiology, Fujita Health University Bantane Hospital, Nagoya, Japan, ⁴Joint Research Laboratory of Advanced Medical Imaging, Fujita Health University School of Medicine, Toyoake, Japan

We hypothesized that compressed sensing (CS) with deep learning reconstruction (DLR) can improve image quality and shorten examination time on not only T2-weighted imaging (T2WI), but also T1-weighted imaging (T1WI) on women’s pelvic MRI, when compared with PI at 1.5T MR system.  The purpose of this study was to compare the utility of CS and DLR for shortening examination time and improving image quality on MRI at 1.5T system in patients with various female pelvic diseases.

Hui Liu¹, Liyun Zheng², Dongmei Wu³, Yongming Dai², and Gaofeng Shi^1
1Department of Radiology, Fourth Hospital of Hebei Medical University, Shijiazhuang, China, ²United Imaging Healthcare, Shanghai, China, ³Shanghai Key Laboratory of Magnetic Resonance, Department of Physics, East China Normal University, Shanghai, China

Clear insights into histological subtype and differentiation grade are essential in cervical carcinoma diagnosis and management. In current study, we demonstrated the feasibility of using high-b-value non-Gaussian diffusion model fractional order calculus (FROC) model to differentiate the tumor subtype and grade of cervical carcinoma. FROC model provided a set of novel diffusion parameters and the combination of these parameters contributed to the best diagnostic performance in differentiation between low-grade and high-grade cervical carcinoma.

Hao Cheng¹, Pu-Yeh Wu², Ming Liu³, Chunmei Li¹, and Min Chen^1
1Department of Radiology, Beijing Hospital, National Center of Gerontology, Institute of Geriatric Medicine, Chinese Academy of Medical Sciences, Beijing, China, ²GE Healthcare, Beijing, China, ³Department of Urology, Beijing Hospital, National Center of Gerontology, Institute of Geriatric Medicine, Chinese Academy of Medical Sciences, Beijing, China

We extracted 104 texture features from relaxation maps of synthetic MRI, DWI and ADC in each of 603 prostate lesions from 297 patients, and compared the diagnostic performance of models based on texture features with mean quantitative values in discriminating PCa from benign lesions and in discriminating csPCa from clinically insignificant lesions. Models of synthetic MRI+DWI+ADC has the highest diagnostic performance in discriminating PCa from SH and noncancerous PZ, and in discriminating csPCa from SH+Gl6 and noncancerous PZ+Gl6. This suggested that texture analysis of synthetic MRI combined with diffusion images can provide better performance in the characterization of prostate lesions.

Qian Tang¹, Weiyin Vivian Liu², Qiqi Zhou³, Yu Xing³, Chao Liu³, and Lin Xu^3
1Biomedical Engineering College, Hubei University of Medicine, Hubei, China, ²GE Healthcare, Beijing, China, ³Department of Radiology,Taihe Hospital, Hubei University of Medicine, Hubei, China

The number of used b values and the b values themselves prolong scan time. Longer scan time deprives patient tolerance, increase more possibility of involuntary movement, and thus limits the application value of IVIM technology. Therefore, we aimed to explore the optimal scheme of rFOV DW imaging with multiple b-values. We found no significant difference of mean ADC, D* and f values but mean D value in our patients between two different b-value schemes, suggesting optimized b-value setting was appropriate in diagnosis of cervical cancer especially for cervical squamous carcinoma.