ISMRM25 - AI-Powered Analysis in Neuroimaging

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

AI-Powered Analysis in Neuroimaging

Digital Poster

AI & Machine Learning

Thursday, 15 May 2025

Exhibition Hall

09:15 - 10:15

Session Number: D-33

No CME/CE Credit

		Computer Number: 17 4285. Improving Low-Angular Resolution Diffusion MRI with 3D Deep Learning: A Model Assessment View Presentation Video N. Tucksinapinunchai, S. Angkurawaranon, R. Boonsuth, U. Yarach Chiang Mai University, Chiang Mai, Thailand Impact: The improved quality and reliability of the diffusion parametric maps produced by our trained 3D-DL model may be advantageous for clinical applications or the investigation of white matter microstructure in various demographics, including transgender individuals, in future research.
		Computer Number: 18 4286. Advancing Brain Morphometry at 7T: A Pilot Study on Epilepsy Patients View Presentation Video L. Bacha, T. Di Noto, P. B. Venkategowda, K. Prabhu, G. F. Piredda, G. Bonanno, J. Slotboom, D. Seiffge, M. Goeldlin, R. Hoepner, S. Vulliemoz, M. Seeck, K. Schindler, M. Baud, J-P Thiran, T. Kober, T. Hilbert, P. A. Liebig, R. Heidemann, R. Wiest, P. Radojewski, B. Maréchal Siemens Healthineers International AG, Lausanne, Switzerland Impact: This work introduces a novel and reliable brain morphometry algorithm that provides detailed structural insight for enhanced clinical decision support in epilepsy care.
		Computer Number: 19 4287. Deep learning of MRI contrast enhancement for mapping cerebral blood volume from single-modal non-contrast scan with Mamba3D-CNN hybrid model View Presentation Video Y. Zhang, A. Cao, V. Rao, J. Guo Columbia University, New York, United States Impact: By accurately estimating cerebral blood volume, this approach eliminates risks associated with gadolinium administration, such as long-term tissue retention. This advancement enables functional imaging for researchers and clinicians, providing a safe and cost-effective alternative for studying and diagnosing neurodegenerative diseases.
		Computer Number: 20 4288. Joint Extraction of Cerebrum and Cerebellum from Lifespan MRIs View Presentation Video L. Wang, Y. Sun, G. Li, W. Lin, L. Wang UNC-Chapel Hill, Chapel Hill, United States Impact: Our method unifies cerebrum and cerebellum extraction, addressing anatomical differences, reducing time complexity, and ensuring accuracy across modalities and ages. This enhances neurological research and clinical diagnostics by enabling precise analysis and monitoring of brain structures.
		Computer Number: 21 4289. Three-Point Deep Learning Framework for Protocol-Independent and AIF-Free DCE-MRI Parameter Estimation in Gliomas View Presentation Video P. Prajapati, A. Kandpal, S. Srivastava, R. Gupta, A. Singh Indian Institute of Technology Delhi, New Delhi, India Impact: The proposed DL approach enables robust DCE-MRI quantification in gliomas using minimal temporal sampling, eliminating AIF dependencies while maintaining accuracy in substantially less time. Facilitating multi-center clinical adoption and efficient pre-operative tumor characterization and treatment monitoring.
		Computer Number: 22 4290. DeepCESTSig: Brain and Tumor Subregion Delineation via Chemical Exchange Saturation Transfer (CEST) MRI View Presentation Video J. R. Rajput, M. S. Fabian, T. A. moehle, A. Mennecke, M. Schmidt, A. Dörfler, A. Maier, M. Zaiss Universitätsklinikum Erlangen, Erlangen, Germany Impact: Segmentation based on CEST signatures improves the delineation of brain tissue and identification of tumors, enabling better clinical decision-making. The approach improves neuroimaging techniques by using biochemical contrasts and can improve the results in the diagnosis of malignant brain areas.
		Computer Number: 23 4291. Task-Agnostic Brain Representations: A Foundation Model for fMRI Using Masked Autoencoders M. Ferrante, S. Iervese, L. Astolfi, N. Toschi University of Rome Tor Vergata, Rome, Italy Impact: We foundation model for fMRI, trained on resting-state data from the HCP to develop generalizable brain representations. Using self-supervised learning, this task-agnostic model can be applied to various neuroscience tasks, including physiological prediction and brain decoding.
		Computer Number: 24 4292. Investigating Prognostic Value of Dynamic Susceptibility Contrast Perfusion MRI-Derived Features for Glioblastoma Survival by Deep Learning View Presentation Video L. Tang, Q. Gu, T. Wu, A. Goldman-Yassen, H. Mao Emory University, Atlanta, United States Impact: Using our Hierarchical Density-Based Network (HDBNet) to investigate hemodynamic information in dynamic susceptibility contrast perfusion-weighted imaging (DSC-PWI) reveals key features that can enhance GBM prognosis, supporting the importance of including hemodynamic and physiological imaging data in future GBM research
		Computer Number: 25 4293. Investigate brain volumes with machine learning algorithms to differentiate PD and PSP patients View Presentation Video C. Calomino, M. Bianco, A. Quattrone Magna Graecia Univerisity, Catanzaro, Italy Impact: This study provides first evidence of alterations in subcortical volume and cortical thickness between Parkinson’s disease patients and Progressive Suplanuclear Palsy patients using a rigorous approach combining nested cross validation, XGBoost and SHAP as feature selection.
		Computer Number: 26 4294. Radiomics features reveal the peritumoral heterogeneity of glioblastoma L. Rui, A. Kai, Z. Jing Lanzhou University Second Hospital, Lanzhou, China Impact: The imaging features can provide objective evidence for the peritumoral heterogeneity of GBM and MT, and provide help for the clinical treatment of patients.
		Computer Number: 27 4295. The effect of time-delayed contrast-enhanced T2-FLAIR on the visualization of large-volume brain metastases S. Du, Y. Yin, D. Pylypenko, G. Gong Shandong Cancer Hospital, Jinan, China Impact: Our results show that combining time-delayed CE T2-FLAIR and CE T1WI enhances BM visualization and GTV segmentation accuracy, allowing quantitative analysis of BM imaging differences in various sequences through GTV volume and shape evaluation.
		Computer Number: 28 4296. BOLD Acquisitions and GAN Synthetized VASO Contrasts for Rapid Layer-dependent fMRI View Presentation Video A. Saxena, D. Bharti, T. B. D. Yeo, A. Ajala GE Healthcare, Bangalore, India Impact: We present a method to eliminate the need for implementing VASO pulse sequence by synthetically generating VASO images from acquired BOLD images.
		Computer Number: 29 4297. Leveraging transfer learning for the super-resolution reconstruction of QSM with limited data for the study of the cerebrovasculature View Presentation Video S. Zappalà, E. Patitucci, I. Driver, D. Gallichan, R. Wise, M. Germuska Cardiff University, Cardiff, United Kingdom Impact: By demonstrating the effectiveness of transfer learning with a 3D Densely Connected Super-Resolution Network (DCSRN) model, this study provides a practical approach for researchers to improve the resolution of their own QSM data, even with limited resources.
		Computer Number: 30 4298. Cross-modal brain decoding: using fMRI to decode video stimuli from integrated sensory streams M. Ferrante, T. Boccato, N. Toschi University of Rome Tor Vergata, Rome, Italy Impact: This work opens pathways for more accurate brain decoding in multisensory contexts, potentially advancing brain-computer interfaces and aiding clinical applications in sensory processing disorders.
		Computer Number: 31 4299. Enhancing 3D Brain MRI Using Super-Resolution through U-Net Architecture View Presentation Video S. Singh, B. V. R. Kumar, S. Pathak, R. Jha, M. Singh, A. Parihar, B. Ojha, C. Srivastava, D. Dwivedi King George's Medical University, Lucknow, India Impact: Utilizing SR for 3D MRI images is uncommon, yet it significantly enhances MRI efficiency and resolution. The proposed architecture reduces computational costs while improving results, facilitating quicker MRI execution without compromising image quality.
		Computer Number: 32 4300. Predictors of Ablation Therapy Response in Moderately Differentiated HCC using Radiomics Features from Multi-phasic DCE-MRI View Presentation Video X. Li, Z. Mohammadigolda, Q. Miao, P. Keshavarz, A. Suri, J. Chiang, K. Sung, D. Lu University of California, Los Angeles, Los Angeles, United States Impact: This study identifies radiomics shape and texture-based features from multiphase MRI that differentiate responders from non-responders to thermal ablation in moderately differentiated HCC. These features offer insights into quantitative biomarkers, emphasizing the role of imaging in optimizing therapy.

Similar Session(s)