ISMRM25 - AI for Diagnosis/Prognosis: Neuro II

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

AI for Diagnosis/Prognosis: Neuro II

Digital Poster

AI & Machine Learning

Monday, 12 May 2025

Exhibition Hall

09:15 - 10:15

Session Number: D-39

No CME/CE Credit

		Computer Number: 1 1531. Brain Connectivity based Prediction of Trait anxiety using Graph Neural Network View Presentation Video S-C Jung, H-J Song, D-H Kim Yonsei University, Seoul, Korea, Republic of Impact: This study suggests the possibility to complement traditional psychological assessments by predicting accurately trait anxiety levels using MR images. The approach could simplify mental health diagnostics, raise new questions about imaging-based predictions and improve access to timely interventions.
		Computer Number: 2 1532. Three-Class Radiomics Models for IDH-Mutation and 1p19q-Codeletion Status Prediction in Adult-type Diffuse Gliomas L. yanhua Chinese people's liberation army (PLA) general hospital, Beijing, China Impact: Three-class MRI radiomics can preoperatively predict IDH and 1p19q-codeletion with satisfied performance, which is helpful for glioma risk stratification.
		Computer Number: 3 1533. Habitat-based MR radiomics enhances the ability to pre-operatively predict tumor consistency of patients with meningioma View Presentation Video G. Tan, J. Zhang, L. Yang, M. Cai, Y. Huang, W. Liu, X. Liu The Affiliated Yuebei People's Hospital of Shantou University Medical College, Shaoguan, Guangdong Province, China, Shaoguan, China Impact: This study firstly implemented habitat analysis in the field of tumor consistency prediction of meningioma. Our results showed the meningioma heterogeneity features not only significantly improved predictive performance, but also provided new insights on tumor proliferation mechanisms of meningiomas.
		Computer Number: 4 1534. Developing a radiomics model to predict tumor consistency of pituitary adenomas using multicenter MRI data View Presentation Video J. Wu, P. Wang, J. Xiao, G. Zada, J. Chen, E. Briseno, Z. Fan University of Southern California, los angeles, United States Impact: This research underscores the importance of applying radiomics to diverse, generalized patient data, advancing its potential for real-world clinical use and demonstrating its adaptability to varied patient and imaging conditions in practical medical scenarios.
		Computer Number: 5 1535. Machine learning based MRI radiomics model in predicting postoperative severe poor outcomes after resection of meningioma. View Presentation Video G. Tan, J. Zhang, L. Yang, M. Cai, Y. Huang, W. Liu, X. Liu The Affiliated Yuebei People's Hospital of Shantou University Medical College, Shaoguan, Guangdong Province, China. , Shaoguan, China Impact: The novel model can non-invasively predict PSPO after meningioma resection, enabling early identification of high-risk patients. This approach can optimize clinical decision-making and enhance postoperative management, ultimately improving outcomes for meningioma patients.
		Computer Number: 6 1536. A U-Net Based Machine Learning Approach with Augmentation for Enhanced Precision in Multiple Sclerosis Lesion Segmentation from Multi-Modal MRI View Presentation Video U. Sakoglu, O. Cetin, B. Canel, G. Dogali University of Houston - Clear Lake, Houston, United States Impact: This study’s robust MS lesion segmentation model could complement and improve diagnostic precision and monitoring for clinicians, leading to personalized treatment insights. It enables researchers to explore further multi-modal MRI benefits and model optimizations, ultimately enhancing patient care and outcomes.
		Computer Number: 7 1537. Prediction of Survival During Immunotherapy of Recurrent High-Grade Glioma Using End-to-End CNN Deep Learning Versus Radiomics Models View Presentation Video G. Young, Q. Wan, C. Lindsay, C. Zhang, J. Kim, X. Chen, J. Li, R. Huang, D. Reardon, L. Qin Brigham and Women's Hospital, Boston, United States Impact: End-to-end CNN models can produce similar accuracy in recurrent HGG patient survival prediction during immunotherapy, compared to robust-feature radiomics from manual segmentation, and may add value in initial patient selection for immunotherapy trials, and personalization of therapy.
		Computer Number: 8 1538. Prediction of progression in Parkinson’s disease based on radiomics in T1-weighted MRI and α‑synuclein in cerebrospinal fluid View Presentation Video X. Zhang, Q. Ren, X. Meng, F. Shi, J. Wu Qilu Hospital of Shandong University, Jinan, China Impact: This study demonstrated the feasibility of selecting specific ROIs in standard T1-weighted MRI to predict the course of PD. Our work confirmed that the five brain regions under investigation will in fact alter as PD progresses, as shown by radiomics.
		Computer Number: 9 1539. Auto-Segmentation for Diffuse Pachymeningeal Enhancement in Patients with Spontaneous Intracranial Hypotension View Presentation Video P-H Su, Y-F Wang, P-Y Wu, S-J Wang, C-F Lu National Yang Ming Chiao Tung University, Taipei, Taiwan Impact: This study developed an auto-segmentation model for diffuse pachymeningeal enhancement on brain MRI in SIH patients. It demonstrates the potential to integrate an automatic detection process to assist in the clinical diagnosis and management of SIH patients.
		Computer Number: 10 1540. Utilization of deep learning-based registration method in Parkinson’s disease diagnostic tool View Presentation Video M. Lee, H. Heo, J. Jo, I. Shin, M. S. Kim, S. J. Chung, S. Y. Kang, S. Song Heuron Co.Ltd., Seoul, Korea, Republic of Impact: By changing the ANTs-based SyN registration method in Heuron IPD with TransMorph, a deep learning-based model, we reduced the registration processing time by approximately 32 times while maintaining the diagnostic performance of Heuron IPD.
		Computer Number: 11 1541. Automated Segmentation of Posterior Cranial Fossa in Spontaneous Intracranial Hypotension View Presentation Video P-Y Wu, Y-F Wang, P-H Su, S-J Wang, C-F Lu Department of Biomedical Imaging and Radiological Sciences, National Yang Ming Chiao Tung University, Taipei, Taiwan Impact: This study built the foundation for the automatic detection of SIH from structural MRI, paving the way for future research to facilitate the diagnosis by quantifying brain sagging signs and enhancing the efficiency and accuracy of clinical management.
		Computer Number: 12 1542. Non-contrast MRI based machine learning and radiomics signature can predict the severity of primary lower limb lymphedema X. Li Beijing Shijitan Hospital Affiliated to Capital Medical University, Beijing, China Impact: All five models performed well in distinguishing between the nonsevere group and the severe group. NCMRI based machine learning radiomics signature can predict the severity of primary lower limb lymphedema.
		Computer Number: 13 1543. Machine learning radiomics for prediction of posterior cranial fossa ependymoma PFA and PFB subgroups View Presentation Video R. Xu, H. Kukun, G. Han, Y. Wang The First Affiliated Hospital of Xinjiang Medical University, Urumqi, China Impact: The proposed machine learning model effectively distinguishes between the molecular subtypes of ependymoma, showing strong performance and enhancing diagnostic accuracy, which is expected to provide valuable insights for clinical decision-making.
		Computer Number: 14 1544. Automating early prediction of cerebral palsy: A transfer learning model for infant MRI analysis Z. Jia, T. Huang, Y. Bian, X. Li, J. Yang Xi'an Jiaotong University, Xi'an, China Impact: This model demonstrates the potential of deep transfer learning for early CP prediction, offering reliable support for early intervention and rehabilitation planning in infants aged 6 months to 2 years, with significant clinical application value.
		Computer Number: 15 1545. Predictive Power of Combined Inflammatory Markers and MRI Features for Glioma Prognosis Using Machine Learning View Presentation Video Y. Liu, Y. Wang, J. Zhu, J. Qian, S. Qin, Y. Hong, S. Sun, F. Chen, Q. Zhang, Q. C. Fu, P. Wang, Q. Lv Department of Medical Imaging, Tongji Hospital, Tongji University School of Medicine, Xincun Road No. 389, Shanghai, China Impact: 1.Key MRI-derived features and inflammatory markers were both used to train a model. 2.The models showed superior predictive performance. 3.The models can be used for distinguishing between high-grade and low-grade gliomas.

Similar Session(s)