Cape Town - 2026 ISMRM-ISMRT Annual Meeting and Exhibition

Radiomics Potpourri

Digital Poster

Analysis Methods

Wednesday, 13 May 2026

Digital Posters Row G

13:40 - 14:35

Session Number: 566-03

No CME/CE Credit

Similar Sessions

This session will cover the development and validation of radio mix approaches applied to diverse body parts and disease areas.

		566-03-001. Stability Analysis of Voxel-wise Radiomics Feature Selection in Contrast-Enhanced MRI for Hepatocellular Carcinoma Xiang Li, Yue-Lang Zhang, Shanshan Jiang The Second Affiliated Hospital of Xi'an Jiaotong University, Xi'an, China Impact: The stability analysis of voxel-wise MRI radiomics, which enables robust feature selection, forms the foundation for non-invasive 3D assessment of hepatocellular carcinoma heterogeneity and its subsequent clinical translation.
		566-03-002. Lung UTE MRI in systemic sclerosis: impact of signal normalisation technique and correlation with cardiopulmonary function Laura Saunders, Muad Almsaaid, Amy Simmons, Mack Caraher, Laurie Smith, Alberto Biancardi, Neil Stewart, Andy Swift, Robin Condliffe, Alexander Rothman, David Kiely, Roger Thompson, Jim Wild The University of Sheffield, Sheffield, United Kingdom Impact: Normalised Ultra-short Echo Time (UTE) lung signal correlates with measures of cardiac and pulmonary function in patients with systemic sclerosis (SSc) and shows potential as a quantitative, radiation-free method of quantifying lung structural changes in patients with SSc.
		566-03-003. Cluster Analysis of Pulmonary UTE MRI Helps to Identify Preserved Ratio Impaired Spirometry Imaging Phenotypes Ziwei Zhang, Li Fan, Yuanyuan Cui The Second Affiliated Hospital of Naval Medical University, SHANGHAI, China Impact: PRISm can be conceptually divided into emphysema-predominant and vasculopathy-predominant phenotypes, potentially guiding the development of future clinical management strategies.
		566-03-004. Multi-sequence MRI Radiomic Model for Non-Invasive Classification of Molecular Subtypes of Posterior Fossa Ependymoma Rui Xu, Pahati Tuxunjiang, Wei Zhao, Hanjiaerbieke Kukun, Jingming Jiang, Yuhui Xiong, Ainikaerjiang Aihemaiti, Sijia LI, Yuchen Liu, Yimuran Subi, Yunling Wang Impact: This non-invasive subtyping could directly inform surgical planning and prognostic counseling, potentially improving clinical decision-making. Future work will validate these findings in multi-center cohorts and explore the model's ability to predict therapeutic response and long-term survival outcomes.
		566-03-005. Predicting Drug-Resistant Epilepsy Using Artificial Intelligence and Neuroimaging Mohamad Nazem-Zadeh, Richard Shek-kwan Chang, Sarah Barnard, Heath Pardoe, Ruben Kuzniecky, Duong Nhu, Deval Metha, Daniel Thom, Zhibin Chen, Zongyuan Ge, Terence O'Brien, Ben Sinclair, Meng Law, Patrick Kwan Monash University, Melbourne, Australia Impact: This study demonstrates that integrating MRI-derived biomarkers with clinical data enables early prediction of drug-resistant epilepsy. These AI-driven prognostic tools could shorten treatment delays, improve therapy selection, and guide personalized interventions, advancing precision medicine in epilepsy care.
		566-03-006. Aerobic Fitness Augments Myelin in Men, But Not Women Alma Davidson, Ariana Olivares, Adriana Dipple, Kellie Hoehing, Alexander Chui, Vincenzo Lauriola, Rochelle Goldsmith, Richard Lipton, Kenny Ye, Roman Fleysher, Michael Lipton Columbia University, New York, United States of America Impact: Myelin may serve as a substrate of fitness-related brain health, even among young individuals, but appears less robust among women. Measurement of MWF can advance understanding of mechanisms linking aerobic fitness to brain health and how it can be maximized.
		566-03-007. Multiparametric MRI-based radiomics for predicting intraoperative blood loss in patients with meningiomas Xinru Deng, Dan Luo, Jiankun Dai, Xinlan Xiao The Second Affiliated Hospital of Nanchang University, Nanchang, Jiangxi, China Impact: This study showed clinical-semantic and multiparametric MRI-based radiomic features can preoperatively predicting intraoperative blood loss in meningiomas patients. The application of clinical-semantic and multiparametric MRI features would be beneficial for guiding the surgical plan for patients with meningiomas.
		566-03-008. Radiomics Model Leveraging AI-Extracted ADC Map Features Predicts Biochemical Recurrence in Advanced Prostate Cancer Huihui Wang, Kexin Wang Peking University First Hospital, Beijing, China Impact: This study validates AI-derived tumor segmentation as an expert-equivalent, scalable alternative for radiomics-based prediction of biochemical recurrence in advanced prostate cancer, enabling reproducible risk stratification and timely intervention to improve patient outcomes.
		566-03-009. Radiomics based on cerebellar 3D T1WI for differentiating patients with Levodopa-induced dyskinesia Yini Chen, Li Ding, Bo Sun, Linyou Wang Taizhou Municipal Hospital, Taizhou, China Impact: The radiomics model extracted from cerebellar gray and white matter effectively differentiates between LID and N-LID patients, revealing the heterogeneity characteristics of LID patients from a novel perspective, thereby significantly improving diagnostic performance and providing auxiliary support for clinical diagnosis.
		566-03-010. Radiomics-Based Models for Non-Invasive Diagnosis and Hypoglycemia Prediction in Pediatric Type 1 Diabetes Zheng Cai, Shaoqing Chen, Lu Han, Zhihan Yan, Kun Liu The Second Affiliated Hospital and Yuying Children’s Hospital of Wenzhou Medical University, Wenzhou, China Impact: This study provides clinicians with a non-invasive tool for both diagnosing type 1 diabetes and proactively identifying children at high risk for inpatient hypoglycemia, enabling timely interventions to improve safety.
		566-03-011. Hierarchical complexity as a discriminant of taxonomical orders in mammalian connectomes Keith Smith, Paola Galdi, Jason Smith, Manuel Blesa Cábez Computer and Information Sciences, Glasgow, United Kingdom Impact: Hierarchical Complexity plays a unique and powerful role in distinguishing connectomes across taxonomical orders. This demonstrates the significance of HC in understanding brain network evolution and opens up new research directions for posing brain network complexity as an evolutionary paradigm.
		566-03-012. Feasibility of using Ultra-Short Echo Time MRI to assess the dynamic change in airway structure during tidal breathing Yves Brown, Neil Stewart, Freya De Monte, Amy Simmons, Laurie Smith, Jim Wild The University of Sheffield, Sheffield, United Kingdom Impact: Understanding the change in airway morphometry during breathing using 3D lung MRI could aid in the targeting of treatments for the airways.
		566-03-013. Estimating Brain Age and Identifying SuperAgers Using Bayesian Neural Networks Carlos Leandro Silva dos Prazeres, Alessandra Goulart, Claudia Leite, Alexandre Chiavegatto Filho, Claudia Suemoto, Isabela Benseñor, Paulo Lotufo, Maria Garcia Otaduy Hospital das Clínicas, Faculdade de Medicina da Universidade de São Paulo, Sao Paulo, Brazil Impact: This work presents an innovative machine learning model to estimate brain age based on simple MRI images. The so called brain-age-gap (BAG) was shown to be sensitive to identify SuperAgers, and can become an important research and clinical biomarker.
		566-03-014. MRI Analysis of Undifferentiated Pleomorphic Sarcoma: Correlating Imaging Features with Histological Grade Junhui Yuan, Shaobo Fang, Xuejun Chen The Affiliated Cancer Hospital of Zhengzhou University & Henan Cancer Hospital, Zhengzhou, China Impact: MRI-derived markers such as growth pattern, necrosis extent, ADC values enable noninvasive grading of undifferentiated pleomorphic sarcoma, improving prognostic assessment and guiding individualized therapy. These findings may stimulate further research into MRI-based radiomics for sarcoma stratification and treatment response prediction.