| Computer # |
| 3753.
 | 49 | Synthetic MRI to evaluate myometrial invasion and predict pathologic type of endometrial cancer: compared with high-resolution T2WI and DWI Yue Li1, Yigang Pei2, Hui Liu2, Weiyin Vivian Liu3, Yijing Luo2, Yu Bai2, Xiaorong Ou2, and Wenzheng Li2 1Radiology, Department of Radiology, Xiangya Hospital, Central South University, Changsha, China, 2Department of Radiology,Xiangya Hospital, Central South University, Changsha, China, 3GE Healthcare,MR, Beijing, China Keywords: Synthetic MR, Cancer Motivation: Depth of myometrial invasion (DMI), lymphovascular invasion (LI) and pathological types of endometrial cancer (EC) affect decision-making on an optimal treatment plan for uterus. Goal(s): This study aimed to investigate the feasibility of MAGIC in evaluating the DMI and predicting the pathological types of EC. Approach: The assessment performance of MAGIC in comparison to hr-T2WI on DMI and prediction of EC types in comparison to ADC were performed. Results: T2 and PD together had a superior predictive performance to ADC only, and sy-T2WI showed the similar assessment performance on DMI to hr-T2WI. Impact: One more application of MAGIC-generated contrast
images and quantitative parameter maps in cervical diagnosis may provide
additional information on diagnosis of DMI and classification of pathological
types of EC. |
| 3754.
 | 50 | Synthesis of 4D Flow MRI Data using Particle-based Bloch Simulations in CMRsim Charles McGrath1, Pietro Dirix1, Jonathan Weine1, and Sebastian Kozerke1 1University and ETH Zurich, Zurich, Switzerland Keywords: Synthetic MR, Simulations Motivation: Signal model-based synthesis of 4D flow MRI data from CFD is promising to study image acquisition, reconstruction methods and augment data for inference. However, the approach assumes idealized encoding, and particle-based Bloch simulations are needed to include complex effects. Goal(s): Implement a particle-based Bloch simulation in CMRsim for generating synthetic multi-venc 4D flow data. Approach: Using pulsatile, stenotic U-bend CFD data as input, simulations of moving magnetization in CMRsim generate synthetic 4D flow MRI data.
Results: 4D flow simulation is feasible in CMRsim, demonstrating tractable simulation times and good agreement with ground truth and magnetization history-based effects, namely acceleration and intra-voxel dephasing.
Impact: Numeric Bloch simulations in CMRsim permit a
detailed study of imaging protocol-dependent flow-related artifacts in 4D flow
MRI of complex flows and can thereby assist in improving acquisition and
reconstruction approaches. |
| 3755.
 | 51 | Predicting the efficacy of induction chemotherapy for nasopharyngeal carcinoma: histogram analysis of quantitative synthetic MRI Junhao Huang1, Jiuquan Zhang1, Huanhuan Ren1, Daihong Liu1, Jing Zhang1, Hong Yu1, Yong Tan1, and Lisha Nie2 1Chongqing University Cancer Hospital, School of Medicine, Chongqing University, Chongqing, China, 2GE HealthCare MR Research, Beijing, China Keywords: Synthetic MR, Quantitative Imaging, nasopharyngeal carcinoma Motivation: In nasopharyngeal carcinoma (NPC) patients, there is a significant variation in tumor response to induction chemotherapy (ICT), which directly impacts prognosis. To address this issue, we aimed to explore a novel imaging biomarker based on pre-treatment synthetic MRI to predict which patients would benefit the most from the additional ICT. Goal(s): Evaluating the value of pre-treatment synthetic MRI quantitative parameter map histogram characteristics in predicting the efficacy of ICT for NPC. Approach: 40 NPC patients were prospectively enrolled, and each was imaged with synthetic MRI. Results: T2_Minimun and T2_RootMeanSquared show promise as imaging markers for predicting the response to ICT in NPC. Impact: The utilization of synthetic MRI may
serve as an effective diagnostic tool for evaluating the response to ICT in
clinical practice. Identifying patients who are unlikely to respond to ICT
early on, can help offer
them alternative treatment options. |
| 3756.
 | 52 | Synthetic quantitative magnetic resonance imaging and quantitative susceptibility mapping to reveal brain function in Type 2 Diabetes Hailing Zhou1, Wenjie Chen1, Shouchao Wei2, Yongsheng Liang1, and Weiyin Vivian Liu3 1Department of Radiology, Central People's Hospital of Zhanjiang, Zhanjiang, China, 2Research Assistant in Department of Clinical Research Institute, Central People's Hospital of Zhanjiang, Zhanjiang, China, 3GE Healthcare, MR Research China, Beijing, China Keywords: Synthetic MR, Quantitative Imaging, synthetic magnetic resonance imaging, quantitative susceptibility mapping, type 2 diabetes, cognitive impairment Motivation: Identification of high-risk dementia in T2DM is very important for early intervention. Goal(s): To explore a feasible imaging approach in early discovering T2DM-driven cognitive impairment. Approach: To compare ROI-based retrieved relaxation time of synthetic magnetic resonance imaging and quantitative susceptibility values between T2DM and healthy cohorts (NCs) and also correlate significant differences of all measurements with Montreal Cognitive Assessment (MoCA) scores. Results: T2DM had higher T1 and T2 relaxation time, PD and QS values in some brain regions than NCs. Moreover, T1 value of left insula was negatively associated with MoCA. Impact: Synthetic MRI and QSM can detect abnormal brain areas associated with in T2DM, and the former had more potential in clinically diagnosing early alteration in T2DM due to more direct visualization and measurements on the scanner console. |
| 3757.
 | 53 | Pushing MP2RAGE boundaries: ultimate time-efficient parameterization combined with exhaustive T1 synthetic contrasts. Blanche Bapst1,2, Aurélien Massire3, Franck Mauconduit1, Vincent Gras1, Nicolas Boulant1, Juliette Dufour4, Benedetta Bodini4, Bruno Stankoff4, Alain Luciani5, and Alexandre Vignaud1 1CEA, NeuroSpin, Gif-sur-Yvette, France, 2Neuroradiology, Henri Mondor University Hospital, Creteil, France, 3Siemens Healthcare SAS, Saint Denis, France, 4Paris Brain Institute, ICM, Sorbonne Université, Paris, France, 5Medical Imaging, Henri Mondor University Hospital, Créteil, France Keywords: Synthetic MR, Brain, Acquisition Methods, Challenges, High-Field MRI MR, Fingerprinting/Synthetic MR, MR Value, Multiple Sclerosis, Neuro, Parallel Transmit & Multiband, Relaxometry Motivation: Redefining MP2RAGE sequence for clinical efficiency. Goal(s): To provide a time-efficient MP2RAGE parameterization with on-demand synthetic T1-weighted contrasts. Approach: Sequence parameters are chosen to minimize idle time while maximizing CNRWM/GM. Synthetic contrasts are derived from T1 map. Experimental validation is carried for 7T brain imaging using plug-and-play parallel transmission. Results: Time-efficient MP2RAGE reduced acquisition time by up to 40% compared to reference, while maintaining contrast quality. Multiple sclerosis patients benefited from enhanced lesion visualization with a 10min, (0.67mm)3 time-efficient protocol. This optimization enabled shorter acquisition times or higher resolution within a given time budget, while providing an exhaustive set of brain T1-w contrasts. Impact: A time-efficient MP2RAGE sequence parameterization is proposed, resulting in faster, higher-resolution 3D T1-weighted brain imaging compared to conventional settings, combined with a complete set of synthetic T1-w contrasts generated online, with meaningful potential in clinical and research practice. |
| 3758.
 | 54 | Synthesizing Missing MRI Sequences Towards Reliable Brain Tumor Segmentation Using Deep Learning Abdulkhalek Al-Fakih1,2, Abdullah Shazly1,2, Abbas Mohammed1,2, Mohammed Elbushnaq1, Meena Makary1,3, and Mohammed A. Al-masni2 1Department of Biomedical Engineering and Systems, Cairo University, Cairo, Egypt, 2Department of Artificial Intelligence, Sejong University, Seoul, Korea, Republic of, 3MGH/HST Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA, United States Keywords: Synthetic MR, Data Analysis, Brain Tumor Segmentation, MR Sequence Synthesis, nnU-net, GANs, Multi Contrast MR, Deep Learning Motivation: Automated and robust segmentation of brain tumors requires multiple MRI sequences and can expedite neuro-oncological clinical trials. Goal(s): Our goal was to develop a deep learning model for brain tumor segmentation, even when some MRI sequences are missing. Approach: We enhanced a GAN with attention modules to synthesize missing sequences and employed an optimized nnU-Net for segmentation using both real and synthesized sequences. Results: The proposed AI-based model significantly improved brain tumor segmentation, with overall Dice scores increasing from 0.688% when FLAIR is missing to 0.873% using synthesized FLAIR derived from T2, and achieving 0.901% with real FLAIR. Impact: The developed two-stage deep learning framework,
comprising synthesis and segmentation, enhances segmentation of brain tumors in
MRI, especially when real sequences are unavailable. This advancement
accelerates clinical trials and reduces manual segmentation time, yielding
promising results. |
| 3759.
 | 55 | Optimization of cervical cord synthetic T1-weighted MRI for enhancing clinical application Simon Schading-Sassenhausen1, Maryam Seif1,2, Nikolaus Weiskopf2,3, and Patrick Freund1,2,4 1Spinal Cord Injury Center, Balgrist University Hospital, Zürich, Switzerland, 2Department of Neurophysics, Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany, 3Felix Bloch Institute for Solid State Physics, Leipzig University, Leipzig, Germany, 4Wellcome Trust Centre for Neuroimaging, University College London, London, United Kingdom Keywords: Synthetic MR, Precision & Accuracy Motivation: MRI acquisition time is a critical factor for patient comfort and motion artifacts. To reduce total acquisition time, synthetic MRI is a flexible solution. Goal(s): To optimize synthetic T1-w (synT1-w) MRI for increasing the accuracy (i.e. difference between synT1-w and MPRAGE) of spinal cord cross-sectional area measurements. Approach: We optimized and validated synT1-w for tracking neurodegeneration in the cervical cord after spinal cord injury and assessed the required sample size for detecting hypothetical treatment effects. Results: Accuracy of synT1-w improved considerably with a minor remaining bias of -0.5% compared to MPRAGE. 13.5% less participants are required when using synT1-w instead of MPRAGE. Impact: Synthetic MRI can help to optimize imaging protocols in clinical trials by reducing acquisition time and the number of required participants. By improving the accuracy of synthetic T1-weighted images, better comparability with different studies using acquired MRI can be achieved. |
| 3760.
 | 56 | Deep learning imaging-based reconstruction improved the image quality of synthetic high b-value DWI for prostate lesion detecting Li Fan1, Xiuxiu Zhou1, Hanxiao Zhang1, Jiankun Dai2, Jie Shi2, Song Jiang1, Lingling Gu1, and Pei Zhang1 1Second Affiliated Hospital of Naval Medical University, Shanghai, China, 2MR Research, GE Healthcare, Beijing, China Keywords: Synthetic MR, Prostate, Deep learning reconstruction, diffusion weighted imaging, cancer Motivation: Synthetic-DWI (SyDWI) at high-b-value, derived from low-b-value DWI, might be beneficial for prostate cancer evaluation due to better conspicuity of lesions. Relative to conventional reconstruction (ConR), a vendor-provided deep learning reconstruction (DLR) has been reported for improving imaging quality in aspects of higher SNR and imaging sharpness. Goal(s): Investigate the impact of DLR on the image quality of SyDWI for prostate lesion detection. Approach: Low-b-value DWI was reconstructed with DLR and ConR, separately. SyDWIs were generated from DWI_DLR and DWI_ConR. The image quality and lesion conspicuity were compared. Results: SyDWI generated from DWI_DLR showed improved image quality and enhanced prostate lesion detection. Impact: The enhancement of prostate lesion detection would be beneficial for clinical examination. |
| 3761.
 | 57 | SyntheticLGE.jl: An Open-Source Toolbox for Retrospective T1 Fitting and Synthetic LGE Image Generation Calder D. Sheagren1,2, Brandon T. T. Tran1,2, Jaykumar H. Patel1,2, Angus Z. Lau1,2, and Graham A. Wright1,2 1Medical Biophysics, University of Toronto, Toronto, ON, Canada, 2Physical Sciences Platform, Sunnybrook Research Institute, Toronto, ON, Canada Keywords: Synthetic MR, Cardiovascular, Late Gadolinium Enhancement, Synthetic MR Motivation: Synthetic late gadolinium enhancement (SynLGE) has been proposed as a technique to quantify cardiac fibrosis from post-contrast T1 mapping. Current SynLGE techniques use site-specific code, limiting clinical adoption and standardization. Goal(s): Develop a software toolbox for SynLGE image generation using retrospective T1* mapping. Approach: An open-source software SyntheticLGE.jl was implemented in Julia and is publicly available on GitHub with two sample MOLLI datasets for software evaluation. Results: SynLGE image generation is feasible for both SSFP and gradient-echo MOLLI imaging. Impact: We hope that SyntheticLGE.jl can enable standardized and reproducible synthetic LGE image generation in simple and challenging clinical scenarios. |
| 3762.
 | 58 | Synthesizing multiple realistic MR phase images using a multi-modal generative model Nikhil Deveshwar1,2, Abhejit Rajagopal1, Michael Lustig2, and Peder E.Z. Larson1 1Radiology and Biomedical Imaging, University of California, San Francisco, San Francisco, CA, United States, 2Electrical Engineering and Computer Science, University of California, Berkeley, Berkeley, CA, United States Keywords: Synthetic MR, Machine Learning/Artificial Intelligence Motivation: Deep learning MRI reconstruction methods face challenges in available datasets to train models. Clinical scans can be a source for diverse data but a challenge is obtaining MRI phase. Goal(s): We propose a method to generate multiple plausible synthetic phase images from a single magnitude-only input. Approach: We train a multi-modal generative model enforcing consistency in the latent space during training. We evaluate the effect of latent vector dimension on diversity and quality of the synthetic images with FID score and training image reconstruction models with this synthetic data. Results: Higher latent vector dimension resulted in more diverse and higher quality synthetic images. Impact: This method could be used to generate multiple plausible phase images from a single scan to model effects of varying field homogeneity, RF coils, echo time, motion, flow, and susceptibility |
| 3763.
 | 59 | The value of synthetic MRI in differentiating metastatic and non-metastatic lymph nodes in squamous cell carcinoma of head and neck Haoran Wei1, Fan Yang1, Xiaoduo Yu1, Lizhi Xie2, and Meng Lin1 1Department of Diagnostic Radiology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China, 2MR Research China, GE Healthcare, Beijing, Beijing, China Keywords: Synthetic MR, Head & Neck/ENT, Differential diagnosis Motivation: The presence of metastatic lymph nodes in HNSCC indicates a worse prognosis, biopsy is invasive and has a high incidence of false-negative results. The diagnosis value of synthetic MRI needs to be studied further. Goal(s): To explore the value of the features derived from synthetic MRI in distinguishing between benign and malignant lymph nodes. Approach: This study included lymph nodes of HNSCC which were confirmed by pathology, and utilized multiple methods to select meaningful features. Results: Features derived from synthetic MRI have satisfactory discrimination efficiency. Impact: Synthetic
MRI may be able to be used as a method to help assemnet the lymph nodes in
clinical practice. |
| 3764.
 | 60 | Radiological Image Quality Assessment of Synthetic 3T MRI Image from 64mT MRI Kh Tohidul Islam1, Shenjun Zhong1,2, Parisa Zakavi1, Helen Kavnoudias3,4, Shawna Farquharson2, Gail Durbridge5, Markus Barth6, Katie L. McMahon7, Paul M. Parizel8,9, Gary F. Egan1, Andrew Dwyer10, Meng Law3,4, and Zhaolin Chen1,11 1Monash Biomedical Imaging, Monash University, Clayton, Victoria, Australia, 2Australian National Imaging Facility, Queensland, Australia, 3Neuroscience, Monash University, Clayton, Victoria, Australia, 4Radiology, Alfred Hospital, Victoria, Australia, 5Herston Imaging Research Facility, University of Queensland, Queensland, Australia, 6School of Electrical Eng. and Computer Science, University of Queensland, Queensland, Australia, 7School of Clinical Science, Queensland University of Technology, Queensland, Australia, 8David Hartley Chair of Radiology, Royal Perth Hospital, Western Australia, Australia, 9Medical School, University of Western Australia, Western Australia, Australia, 10South Australian Health and Medical Research Institute, South Australia, Australia, 11Data Science and AI, Monash University, Clayton, Victoria, Australia Keywords: Synthetic MR, Low-Field MRI Motivation: The necessity to enhance the quality of portable low-field MRI images, which are crucial for wider accessibility but lack high-resolution, drives this research. Goal(s): This study aims to determine whether Synthetic 3T technology can elevate low-field image quality to that of high-field 3T standards, making it diagnostically adequate. Approach: We employed a generative network to transform low-field images to higher quality, maintaining pixel-level accuracy and structural integrity. The study involved a paired dataset from 37-healthy subjects and an evaluation on 20-images by two neuroradiologists. Results: Synthetic 3T demonstrated improved clarity, structure, and contrast, aligning closer to the 3T-quality than the original low-field images. Impact: This investigation highlights the potential of Synthetic 3T to bridge the gap in portable MRI imaging, enabling broader clinical utility. Further research could pivot on its application in pathological cases, with an aim to enhance diagnostic capabilities in resource-limited settings. |
| 3765.
 | 61 | Investigating Myelin Abnormalities in Major Depressive Disorder: Insights from Synthetic Magnetic Resonance Imaging Junyan Wen1, Zhimin Chen1, Liya Gong1, Wei Cui2, Liaoming Gao1, Ziqi Wu1, Shanshan Yang1, Yanyu Hao1, and Ge Wen1 1Department of Medical Imaging, Nanfang Hospital, Guangzhou, China, 2MR Research China, GE Healthcare, Beijing, China Keywords: Quantitative Imaging, Quantitative Imaging, Major depressive disorder; synthetic MRI; myelin Motivation: Major depressive disorder is a common mental illness with biological processes, such as myelin alterations, that are not yet fully understood. Goal(s): This study aims to explore myelin abnormalities in MDD patients utilizing Synthetic MRI (SyMRI) technique. Approach: Myelin content measured by SyMRI was compared between 32 MDD patients and 51 healthy controls, and the study further investigated the relationship between myelin alterations and the depressive symptoms. Results: Several white matter regions exhibited a significant decrease in myelin content in individuals with MDD, and a correlation was found between the depressive symptoms and these myelin deficits. Impact: The present study highlights
the potential of Synthetic MRI technique as a valuable tool for measuring
myelin content, which could play a crucial role in understanding the underlying
neurobiological mechanisms of Major Depressive Disorder. |
| 3766.
 | 62 | Assessing the Diagnostic Potential of Synthetic MRI for Hypogonadotropic Hypogonadism in Thalassemia Patients Meicheng Li1, Hongxiu Zeng1, wei cui2, Cheng Tang1, and Peng Peng1 1Department of Radiology, The First Affiliated Hospital of Guangxi Medical University, Nanning, China, 2GE Healthcare, MR Research China, Beijing, China Keywords: Hematology, Oncology, Quantitative Imaging, thalassemia; hypogonadotropic hypogonadism; synthetic MRI Motivation: Pituitary T1 values and pituitary height have potential as predictive markers for hypogonadotropic hypogonadism (HH) in thalassemia major (TM) patients, yet their assessment typically needs multiple MRI scans. Goal(s): Assess the diagnostic potential of Synthetic MRI in detection of HH, with the advantage of obtaining the required MR measurements through a single scan. Approach: Pituitary T1, T2 and pituitary height were measured in 112 TM patients using SyMRI technique. Results: The AUC values for diagnosing HH with pituitary T1 values and pituitary height were 0.736 and 0.753, respectively, and the AUC value of combining these two measurements was 0.813. Impact: Synthetic MRI technology facilitates the diagnosis of
hypogonadotropic hypogonadism in TM patients, and the combination of pituitary
T1 values and pituitary height yields high diagnostic accuracy for
hypogonadotropic hypogonadism in TM patient. |
| 3767.
 | 63 | DeepDenoiseNet: a convolutional neural network trained with synthesized images from inversion-recovery maps for SASHA denoising Xiaofeng Qian1, Ancong Wang1, Yingwei Fan1, Yafeng Li2, Bowei Liu3, Yongsheng Jin4, Haiyan Ding3, and Rui Guo1 1Shool of Medical Technology, Beijing Institute of Technology, Beijing, China, 2China Electronics Harvest Technology Co.,Ltd, Beijing, China, 3Center for Biomedical Imaging Research, Department of Biomedical Engineering, School of Medicine, Tsinghua University, Beijing, China, 4Department of Infectious Diseases, The Affiliated Hospital of Yan’an University, Yan’an, Shanxi, China Keywords: Myocardium, Myocardium, SASHA Motivation: SASHA T1 has high accuracy but low precision due to the low SNR of T1-weighted images. Convolutional neural network has the potential to improve SASHA T1 precision by using spatio-temporal correlations. Goal(s): The aim of this study is to develope a convolutional neural network for improving SASHA T1 precision. Approach: We implemented a convolutional neural network (DeepDenoiseNet) and trained it using synthesized SASHA images from co-registered high-quality T1, T2, and M0 images. Different-level noise was added to simulate low SNR SASHA images. Results: DeepDenoiseNet could reduce the impaction from noise and improve SASHA T1 precision. Impact: The deep convolutional neural network trained with synthesized images and simulated noise could improve SASHA T1 precision. |
| 3768.
 | 64 | An Inpainting-based Method for MRI Synthesis Zhihao Zhang1, Long Wang2, Lei Xiang1, Ryan Chamberlain2, Enhao Gong2, XiaoEr Wei3, Xinyu Song4, and Yuehua Li3 1SubtleMedical, Shanghai, China, 2SubtleMedical, Menlo Park, CA, United States, 3Institute of Diagnostic and Interventional Radiology, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, China, 4Department of Radiology Intervention, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, China Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence, Image Synthesis, Inpainting-based Motivation: The existing deep learning-based MRI sequence synthesis methods are prone to obscure or simulate pathology.
Goal(s): To develop an inpainting-based image synthesis network (IBSNet) for MRI sequence high-fidelity synthesis. Approach: We designed a dual-branch structure to focus on global and local information extraction. Moreover, a joint loss is proposed to constrain the network from signal intensity, structural similarity, and edge preservation. An attention module is used to refine the intermediate feature maps from both the channel and spatial dimensions. Results: The results show that our method outperforms other methods based on the encoder-decoder network, generative adversarial network (GAN), and diffusion model. Impact: Our proposed inpainting-based image synthesis network can generate the target sequence from existing sequences, which can reduce the scanning time of MRIs and improve the patient experience. |