Cape Town - 2026 ISMRM-ISMRT Annual Meeting and Exhibition

Classification: Other Cancer

Digital Poster

Analysis Methods

Thursday, 14 May 2026

Digital Posters Row C

13:40 - 14:35

Session Number: 662-03

No CME/CE Credit

Similar Sessions

The session covers applications of classification methods on cancer MRI.

Skill Level: Basic,Intermediate,Advanced

		662-03-001. Differentiation of IDH mutant grade 4 astrocytoma from IDH wild type glioblastoma using ITSS volume from SWI-MRI Satyajit Maurya, Sanskriti Srivastava, Ewunate Kassaw, Anup Singh Indian Institute of Technology, Delhi, India Impact: The proposed method can aid radiologists in non-invasively determining the IDH mutation status in grade 4 gliomas using SWI with a high degree of accuracy, achieving pre-operative, objective and accurate tumor assessment.
		662-03-002. Lumbar Vertebral Marrow PDFF Improves the Diagnostic Specificity for Prostate Cancer: a Preliminary Study Qinglin Xie, Xueting Du, Qianyi Qiu, Yi Yang, Xinru Zhang, Yiou Wang, Jinling wu, Meimei Zhang, Yuting Ling, Xiaoyun Liang, Zhongping Zhang, Peng Wu, Ge Wen, Xiaodong Zhang The Third Affiliated Hospital Southern Medical University, Guangzhou, China Impact: Lumbar vertebral marrow PDFF complements conventional imaging and biomarkers, improving diagnostic specificity and potentially reducing unnecessary prostate biopsies for patients with suspected prostate cancer.
		662-03-003. Value of Multimodal MRI-Based Radiomics in Predicting Targeted Therapy Efficacy for Locally Advanced Nasopharyngeal Carcinoma Zijing Lin, Zhiqiang Chen, Guangxu Han The First Affiliated Hospital of Hainan Medical College‌, Hainan, China Impact: The multimodal MRI radiomics model shows promise in predicting targeted therapy outcomes for locally advanced nasopharyngeal carcinoma, potentially aiding in personalized treatment strategies and improving patient management through better preoperative risk assessment.
		662-03-004. 3D MRI-Based Breast Cancer Classification: Leveraging Segmentation-Guided Multi-Modality Fusion Kaiting Wang, Pengchen Liang, Zhifeng Chen, Xiaoyun Liang, Darong Zhu, Shiwei Wang Zhejiang Chinese Medical University, Hangzhou, China Impact: This lightweight, robust 3D CNN pipeline automates tumor segmentation and classification, enabling streamlined clinical workflows and enhanced diagnostic confidence in breast cancer screening.
		662-03-005. Nomogram for Preoperative LNM Prediction in Rectal Cancer: Integrating DCE-MRI-Derived ITH Score and Clinical Features Yun Sun, Gang Huang, Kai AI Gansu University of Chinese Medicine, Lanzhou, China Impact: Enables personalized preoperative LNM risk stratification in RC, guiding tailored treatment decisions and improving prognosis.
		662-03-006. Time-Dependent Diffusion MRI of Papillary Thyroid Carcinoma for Predicting Central and Occult Cervical Lymph Node Metastasis Wen Zhao, Zongfang Li, Bo He, Yida Yin, Yu Zhang, Meining Chen, Wei Sheng, Thorsten Feiweier The First Affiliated Hospital of Kunming Medical University, Kunming, China Impact: Microstructural metrics from td-dMRI of papillary thyroid carcinoma enable noninvasive prediction of central and occult lymph node metastasis, offering valuable biomarkers to guide surgical planning, improve detection, and potentially enhance prognosis in thyroid cancer patients.
		662-03-007. Voxel-wise DCE Heterogeneity Index (HI) map for Evaluating Tumor Homogenization and Vascular Normalization in Breast Cancer Priyadharshini Babu, Mythili A, Anandh Ramaniharan Vellore institute of technology, Vellore, India Impact: The proposed voxel-wise DCE-HI map quantifies intra-tumoral heterogeneity dynamics during therapy. Consistent decrease in kurtosis with minor entropy variations in responders indicates vascular normalization and therapy-induced homogenization, offering a simple imaging signature for early treatment response assessment in breast cancer.
		662-03-008. Multi-Modal Non-Contrast Quantitative MRI for Accurate Identification of High-Grade Gliomas: A Feasibility Study Shiwei Lai Guiqian International Hospital, guiyang, China Impact: Multi-modal non-contrast quantitative MRI enables accurate HGG identification, eliminating GBCA dependency and improving accessibility for high-risk patients.
		662-03-009. Multiparametric Histogram Analysis for the Differentiation of IDH-mutant grade-4 Astrocytomas from IDH-wild-type Glioblastoma Archith Rajan, Nishanth Suri, Ajay Kumar, Stephen Bagley, Richard Phillips, Lisa Desiderio, Steven Brem, Laurie Loevner, Suyash Mohan, Sanjeev Chawla University of Pennsylvania, Philadelphia, United States of America Impact: Noninvasive differentiation of IDH-mutant grade-4 astrocytomas and IDH-wild-type glioblastomas (GBMs) was achieved through multiparametric histogram analysis of diffusion and perfusion MRI. This approach successfully captured subtle microstructural and hemodynamic heterogeneity present within the tumors, allowing for improved diagnostic accuracy.
		662-03-010. Diffusion Kurtosis Imaging-based Multi-level Fusion Deep Learning Model in Histological Subtyping of Cervical Cancer Mandi Wang, Xuemei Zhao, Youcai Wei, Jingshan Gong, Elaine Yuen Phin Lee Shenzhen People's Hospital, ShenZhen, China Impact: The proposed DKI-based CNN classification model with feature level and decision level fusions demonstrated excellent performance in differentiating histological subtypes, suggesting the potential clinicopathological value of DKI in cervical cancer.
		662-03-011. Optimizing Early Breast Cancer Diagnosis: Radiographers’ Readiness for AI-Enhanced Breast MRI in Kenya Lydia Mong'are JOMO KENYATTA UNIVERSITY OF AGRICULTURE AND TECHNOLOGY, NAIROBI, Kenya Impact: This study points out that there are readiness gaps in radiographers’ readiness to implement AI in breast MRI and that this can inform training and resource development.
		662-03-012. APTw subregion radiomics for preoperative evaluation of tumor budding and prognostic stratification in rectal cancer Li Zhang, Kai AI Shaanxi Provincial People's Hospital, Xi'An, China Impact: The APTw subregion radiomics model may serve as a surrogate for Bd 3 and thus a tool for diagnosis and risk stratification prior to surgery.
		662-03-013. Prediction of Lymphatic Metastatic Risk in Rectal Cancer Using MRI-DKI Habitat Radiomic Features and Clinical Immune Markers Leping Peng, Rong Li, Ke Hai, Fan Zhang, Fang Ma, Xiuling Zhang, Kai AI, Lili Wang Gansu University of Chinese Medicine, Lanzhou, China Impact: RC patients who are LNM-negative but LVI-positive, treatment should mirrors that for LNM-positive cases. Tumors' subregions may exhibit distinct patterns of invasion. Combined model provides a tool for predicting high-risk areas of LMR, avoiding the risks associated with pathological detection.
		662-03-014. Optimizing Frequency Selection in Breast MR Elastography: A Prospective Study on Biomarkers and Molecular Subtype Xiaowen Ma, Yifeng Chen, Qin Xiao, Yajia Gu Fudan University Shanghai Cancer Center, Shanghai, China Impact: This pioneering study identifies the optimal frequency for breast MR elastography, establishing key standards for future research and paving the way for multicenter applications, while also validating its utility in characterizing breast lesion biomarkers and molecular subtypes.
		662-03-015. Preoperative Ki-67 Prediction in Invasive Breast Cancer via Adaptive Machine Learning and Multiparametric MRI Radiomics Da Shi, Yanan Shi, Qingyun Wang, Yongzhou Xu, Zhaoyong Li Dongguan Key Laboratory of Radiology and Molecular Imaging, Dongguan, China Impact: Our automated machine learning pipeline noninvasively predicts Ki-67 status in breast cancer, offering a promising imaging biomarker to guide personalized neoadjuvant therapy and reduce unnecessary treatment in patients with low-proliferation tumors.