Cape Town - 2026 ISMRM-ISMRT Annual Meeting and Exhibition - Advanced Tumor Imaging: Bridging Research and Clinical Practice

Advanced Tumor Imaging: Bridging Research and Clinical Practice

Oral

Neuro B

Monday, 11 May 2026

Auditorium 1

16:10 - 18:00

Moderators: Koji Yamashita & Gilbert Hangel

Session Number: 303-03

No CME/CE Credit

Similar Sessions

This session focuses on advanced neuroimaging approaches for tumor characterization, with a particular emphasis on imaging biomarkers and their role in precision oncology. Through state-of-the-art MRI techniques, multimodal imaging, and emerging quantitative and AI-based biomarkers, the session will explore how imaging contributes to diagnosis, prognostication, treatment stratification, and response assessment in brain tumors. Particular attention will be paid to the translational bridge between research developments and clinical practice, highlighting current challenges, validation issues, and future perspectives for imaging biomarkers in neuro-oncology.

Skill Level: Intermediate

16:10		303-03-001. Gadolinium-free MRI using artificial intelligence in glioma: a clinically-oriented benchmark study Ivar Wamelink, Rajeev Essed, Stefan de Vries, Ellis Donders, Frederik Barkhof, Alle Meije Wink, Vera Keil, Stefano Trebeschi Amsterdam University Medical Center, University of Amsterdam, Amsterdam, Netherlands Impact: While GBCA-free segmentation algorithms outperform synthesis algorithms in enhancement detection, prognostic assessment, and treatment response evaluation, both methods remain limited for clinical use and should focus more on clinical applicability than visual quality to increase technology readiness level.
16:21		303-03-002. MRI-based deep learning system for noninvasive neuropathological profiling of adult-type diffuse glioma Yangyang Li, Xiaoming Hong, Chenghao Liu, Junjie Li, Haowen Pang, Siyao Xu, Renlong Zhang, xianchang zhang, Zhizheng Zhuo, Chuyang Ye, Yaou Liu Beijing Tiantan Hospital, Capital Medical University, Beijing, China Impact: MRI-based glioma neuropathology prediction (MRI-GNP) deep learning system is a robust and generalizable tool for preoperative neuropathology markers prediction, which has strong potential to enhance precision diagnostics and support clinical decision-making.
16:32		303-03-003. Quantitative MRI–based Prediction of Contrast Enhancement in Brain Tumors using Deep Learning Dennis Thomas, Sarmad Khan, Mariem Ghazouani, Andrei Manzhurtsev, Seyma Alcicek, Ulrich Pilatus, Elke Hattingen, Florian Buettner, Katharina Wenger Goethe University, University Hospital Frankfurt, Institute of Neuroradiology and Cooperative Brain Imaging Center - CoBIC, Frankfurt am Main, Germany Impact: The proposed Quantitative (q) MRI-based deep learning model predicting contrast enhancement without Gadolinium achieved a mean Dice score of 0.60. The 12-minute Gadolinium-free qMRI protocol, generating T1, T2*, QSM, and PD maps, could replace conventional CE imaging in the clinic.
16:43		303-03-004. A Deep Learning Framework for Rapid Post-Processing of Oscillating Gradient Spin Echo in Glioma Molecular Subtyping Yuhui Xiong, Lizhi Xie, Xin Ge, Ruicheng Ba, Jing Zhang, Bing Wu GE HealthCare MR Research, Beijing, China Impact: The proposed deep learning-based framework democratizes advanced cellular microstructural mapping by enabling fast and accurate OGSE analysis, thereby ameliorating its primary post-processing challenge and significantly enhancing its potential for routine clinical application in glioma diagnosis.
16:54		303-03-005. Leveraging advanced diffusion measures to spatially predict changes in tumor burden during stereotactic ablative radiotherapy David Hormuth, II, Jill De Vis, Yen-peng Liao, Yunxiang Li, Robert Timmerman, Zabi Wardak, Tu Dan, Michael Dohopolski, Xin Cai, Caroline Chung, Thomas Yankeelov, You Zhang, Jie Deng The University of Texas at Austin, Austin, United States of America Impact: Advances in MR-LINAC sequence development, image analysis, and mathematical modeling enables the development of digital twins that characterize changes in tumor cellularity and perfusion during radiotherapy. These patient-specific digital twins could serve as decision support tools for radiotherapy planning.
17:05		303-03-006. Arterial Spin Labeling Reveals Hyperperfusion Linked to Cognitive Performance in Treated Glioma Survivors. Joppe Van Rumst, Laurien De Roeck, Ahmed Radwan, Charlotte Sleurs, Sabine Deprez, Stefan Sunaert, Maarten Lambrecht KU Leuven, Leuven, Belgium Impact: This study demonstrates that ASL reveals tumor-/treatment-related hyperperfusion in glioma survivors, identifying neuro-vascular dysregulation as a potential involved mechanism in cognitive dysfunctioning and highlighting the potential importance of sparing hubs and perfusion-defined regions during radiotherapy to mitigate cognitive decline.
17:16		303-03-007. Radiopathomic Map Profiles of Cell Density Near Contrast Enhancement Predicts Overall Survival In Glioblastoma Benjamin Chao, Samuel Bobholz, Savannah Duenweg, Aleksandra Winiarz, Biprojit Nath, Allison Lowman, Daniel Kim, Mrina Mtenga, Hope Reecher, Michael Barrett, Fitzgerald Kyereme, Adam Lahrache, Elaine Tanhehco, Jennifer Connelly, Max Krucoff, Edward Mrachek, Jamie Jacobsohn, Mohit Agarwal, Rupen Desai, Jennifer Tuscher, Peter LaViolette Medical College of Wisconsin, Wauwatosa, United States of America Impact: Spatial patterns of radio-pathomic cell density maps derived from pre-surgical imaging predict overall survival in patients with glioblastoma. These spatial patterns suggest hyper-cellular infiltrative patterns outside the contrast-enhancing mass, which are investigated here at scale in large openly available datasets.
17:27		303-03-008. Classification of Glioma Subtypes Using 7T MR Spectroscopic Imaging Sara Huskic, Philipp Lazen, Thomas Roetzer-Pejrimovsky, Rebeka Rumbak, Ahmet Azgın, Sagar Acharya, Stanislav Motyka, Bernhard Strasser, Lukas Hingerl, Anita Kloss-Brandstätter, Matthias Preusser, Johannes Leitner, Juliane Hennenberg, Karl Rössler, Guenther Grabner, Wolfgang Bogner, Georg Widhalm, Gilbert Hangel Medical University of Vienna, Vienna, Austria Impact: We show that 7T MRSI provides metabolic imaging markers that accurately classify gliomas into subtypes. This demonstrates the potential of 7T MRSI to support preoperative differential diagnosis and inform treatment decisions, motivating the clinical translation of advanced spectroscopic imaging.
17:38		303-03-009. Tertiary Lymphoid Structures in Glioblastoma: Association with Multiparametric MRI Phenotypic Features and Patient Survival Qing Zhou, Jinlin Zhou Lanzhou University Second Hospital, Lanzhou, China Impact: Glioblastoma (GBM) has a poor, heterogeneous prognosis. This study identifies tertiary lymphoid structures (TLS) as a significant positive prognostic biomarker. The TLS-associated survival benefit is linked to preoperative VASARI MRI features, leading to the development of an MRI-based predictive model.
17:49		303-03-010. Evaluating the Contrast to Noise Ratio in Brain Tumor DSC-MRI Data Collected with the Consensus Low Flip Angle Protocol Alia Khaled, Mahsa Servati, Aliya Anil, Rania Mohamed, Ashley Stokes, John Karis, Melissa Prah, Leland Hu, Jerrold Boxerman, Kathleen Schmainda, C. Chad Quarles The University of Texas MD Anderson Cancer Center, Houston, United States of America Impact: For robust brain tumor DSC-MRI CBV mapping, voxel-wise ΔR2*-time series CNR should exceed 4. In clinical data, low and moderate flip angle acquisitions achieved this in 83% and 85% of voxels, respectively, supporting the clinical utility of the consensus protocol.