ISMRM25 - Generating Synthetic Imaging Data: Part I

ISMRM & ISMRT Annual Meeting & Exhibition • 10-15 May 2025 • Honolulu, Hawai'i

Generating Synthetic Imaging Data: Part I

Digital Poster

AI & Machine Learning

Wednesday, 14 May 2025

Exhibition Hall

08:15 - 09:15

Session Number: D-30

No CME/CE Credit

		Computer Number: 33 3216. MegaNet: Physical Intelligent Metabolite Quantification from Edited Magnetic Resonance Spectra with Synthetic Data Learning View Presentation Video J. Zhang, Z. Tu, Y. Li, N. Gao, X. Jiang, Q. Xu, X. Song, D. Guo, X. Qu Xiamen University, Xiamen, China Impact: It shows that synthetic data learning with physical prior knowledge is probably a reliable method to address the problem that AI training of magnetic resonance spectroscopy (MRS) lacks abundant high-quality data.
		Computer Number: 34 3217. Cerebrovascular Reactivity Measurements using Simultaneous 15O-Water PET and Deep-Learning Synthesized Multi-PLD PCASL View Presentation Video B. Ho, D. Kim, A. Kumar, M. Zhao, A. Fan, Y. Jung, G. Zaharchuk Stanford University, Stanford, United States Impact: Though single-PLD ASL offers shorter scan durations, it largely underestimates CVR. Using deep learning models to synthesize multi-PLD ASL even from a single PLD may allow for more robust CVR estimates while potentially matching the shorter scan times.
		Computer Number: 35 3218. A Generative Model of Cortical Surfaces View Presentation Video C. Tsai, J. Zhao, G. Lin, S. Ahmad, P-T Yap University of North Carolina at Chapel Hill, Chapel Hill, United States Impact: This cortical surface generative model can produce a large number of cortical surfaces for training deep learning models and conducting neuroimaging studies.
		Computer Number: 36 3219. Tractography from T1-Weighted MRI Using Segmentation-Guided and Patch-Based Diffusion Model View Presentation Video D. Sheng, J. Wang, G. Xie, L. J O’Donnell, L. Zhang, F. Zhang University of Electronic Science and Technology of China, Chengdu, China Impact: This study introduces a novel approach for generating tractography directly from T1-weighted MRI rapidly and accurately. This can be a useful tool to facilitate large-scale brain connectivity studies without the resource constraints associated with dMRI.
		Computer Number: 37 3220. High Resolution TSE Image Synthesis Using Denoising Diffusion Models Trained on 7T Image Pairs: Application for Hippocampal Subfield Analysis View Presentation Video J. Li, A. Sajewski, T. Santini, T. Ibrahim University of Pittsburgh, Pittsburgh, United States Impact: This work introduces an alternative solution for salvaging motion-corrupted TSE images, potentially reducing patient exclusion rates and improving the statistical power of neuroimaging studies through diffusion model based image translation.
		Computer Number: 38 3221. Cross-Contrast Enhancement of Brain Tumor Datasets using Deep Learning View Presentation Video R. Griffin, M. Murphy, H. Neeli, E. Leiss, A. Webb, D. Martin, N. Tsekos, P. Martin University of Houston, Houston, United States Impact: Leveraging complementary information from another contrast helps overcome the image fidelity lost at higher acceleration factors, allowing for faster diagnostically useful scanning. We extend the analysis of multi-contrast MRI enhancement to patients with tumors, increasing clinical relevance of results.
		Computer Number: 39 3222. COCONET: A Coordinate-Convolutional patch-based ResUnet for MR Pediatric Image Synthesis View Presentation Video T. Wang, Y. Chen, P. Commean, C. Eldeniz, C. Merrill, G. Skolnick, K. Patel, H. An Washington University in St. Louis, St. Louis, United States Impact: This study provides high-resolution pCT images from pediatric MRI imaging. It provides an alternative pediatric cranial bone imaging method free from ionizing radiation.
		Computer Number: 40 3223. An Adaptive Dual Degradation Diffusion Model for Controllable 5T SWI Synthesis from 3T MRI View Presentation Video Y. An, C. Liu, W. Zhao, Y. Dong, Y. Jiang, X. Song, D. Liang, H. Zheng, Z. Hu Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China Impact: The method has shown effectiveness in enhancing SWI images, providing detailed results for brain microbleed evaluation and neurological assessments. These findings indicate that the proposed model has significant potential for advancing clinical applications in brain-related research.
		Computer Number: 41 3224. Synthesis of Three-dimensional (3D) Multi-contrast Brain Tumor Images with Controlled Tumor Properties Using Generative Models View Presentation Video K. Aki, Y. Takeishi, Y. Jitsumatsu, S. Kuhara, J. Takeuchi, H. Takeshima Kyushu University, Fukuoka, Japan Impact: From non-tumor MR images, images with tumors at controlled locations can be generated using 3D ellipsoids as desirable regions of the tumors. Potential applications include training of machine learning models.
		Computer Number: 42 3225. Dual-Branch Score-Based Diffusion Model Incorporating Radiomics Into Synthesizing PET From MR Images B. Zhang, T. Xie, Z. Cui, X. Dong, J. Lyu, S. Zhang, H. Li, H. Wang, D. Liang, Y. Zhou Shenzhen Institute of Advanced Technology, Shenzhen, China Impact: The proposed method integrates radiomics into the synthesis process, enhancing the clinical utility of MRI-based PET alternatives by producing reliable, radiation-free PET-like images with improved diagnostic fidelity and accessibility.
		Computer Number: 43 3226. Magnetic Resonance Fingerprinting (MRF)-based Virtual Contrast-enhanced MRI Synthesis using Deep Transfer Learning View Presentation Video Y. Ni, J. Lai, C. Liu, W. Li, X. Wang, P. Wang, G. Ren, J. Cai, T. Li The Hong Kong Polytechnic University, Hong Kong, Hong Kong Impact: MRF-based VCE-MRI has the potential to enhance patient safety and streamline clinical workflows by eliminating the need for contrast agents. Offering comparable synthetic accuracy to conventional T1w/T2w-based models, the MRF-based approach features a more efficient single-sequence acquisition.
		Computer Number: 44 3227. Cross-Modal Text Prompts Specified MRI-to-PET Dynamic Image Translation View Presentation Video Y. Jiang, Y. Jin, H. Liu, Y. An, N. Zhang, H. Zheng, D. Liang, J. Liu, R. Chen, Z. Hu Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China Impact: This approach offers a safer imaging alternative by generating synthesis PET images without radioactive tracers, supporting disease diagnosis and monitoring with reduced patient risk. This development could broaden MRI’s clinical application, fostering multi-tracer insights in resource-limited settings.
		Computer Number: 45 3228. CW-GAN: Controllable-Weighting Generative Adversarial Networks for Cross-Domain Multi-Contrast MR Image Synthesis H. Zheng, Z. Wu, G. Wang, C. Cai, S. Cai, Z. Chen Xiamen University, Xiamen, China Impact: CW-GAN provides a powerful tool for MRI image synthesis with controlled weighting, outperforming existing methods in generalization and controllability, offering valuable potential for clinical applications and advancing MRI deep learning-based tasks.
		Computer Number: 46 3229. MRI-Driven Diffusion Model of 18F-FP-CIT PET Image Synthesis for Parkinson's Disease View Presentation Video C. Liu, Y. Jin, H. Liu, J. Han, F. Fu, B. Li, N. Zhang, H. Zheng, D. Liang, Z. Hu Research Center for Medical AI, Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China Impact: This development is valuable for the diagnosis of PD in patients, offering a potential tool to enhance diagnostic capabilities.
		Computer Number: 47 3230. AV1451 Tau PET Synthesis from T1-weighted MR Images with U-NET View Presentation Video F. Vega Lara, M. MacDonald University Medical Center Groningen, Groningen, Netherlands Impact: Our model proves the feasibility of synthesizing AV-1451 PET images from T1-weighted MRI, an approach that could enhance the accessibility for large cohort-studies and early dementia detection, while also reducing costs, invasiveness, and enhancing patient safety by limiting radiation exposure.
		Computer Number: 48 3231. Imputing Longitudinal Infant Brain MRI Features Using a Self-Supervised Transformer Model View Presentation Video C. Ning, R. Zhao, Y. Chen, H. Xu, M. Li, T. Zheng, X. Xu, R. Chen, Y. Zhang, L. Zhao, D. Wu Key Laboratory for Biomedical Engineering of Ministry of Education, Department of Biomedical Engineering, College of Biomedical Engineering & Instrument Science, Zhejiang University, Hangzhou, China Impact: The proposed model is able to impute the missing data in longitudinal studies of infants, which may enrich the information along development trajectory and downstream analyses.

Similar Session(s)