Classification and Analysis in the Body
Digital Poster
Analysis Methods
Wednesday, 13 May 2026
This session focuses on image analysis and tools using data processing and AI methods to classify the body.
|
561-02-001.
Evaluation of Three-Dimensional Double-Echo Steady-State MRI in Characterizing Acetabular Labral Tears
Impact: 3D DESS MRI demonstrated high diagnostic accuracy in
classifying acetabular labral tears based on the MAHORN classification, showing
good agreement with arthroscopic findings. And the clock-face method is
recommended as a reliable technique for the precise localization of labral
tears.
|
||
|
561-02-002.
Influence of Site Effects on Radiomics-Based Knee Injury Diagnosis
Impact: Magnetic field strength (MFS) substantially affects feature
distributions and may introduce bias into machine learning-based disease
classification. These findings highlight the need to balance field strength
variations in multi-site MRI datasets to enhance model generalization and
clinical applicability.
|
||
|
561-02-003.
Multimodal MRI-Based Deep Learning for Automated Knee Cartilage Injury Classification
Impact: This multimodal framework provides a robust, standardized decision-support tool for knee cartilage injury diagnosis, with potential to enhance clinical management of early osteoarthritis.
|
||
|
561-02-004.
Application of Uniform Connection Net Based on Multi Hypergraph Dynamic Node Framework for Carotid Plaque Vulnerability
Impact: Enhances stroke risk prediction for clinicians, offers a scalable, robust solution, UniConnNet, for scientists to explore, and helps patients get timely interventions by accurate carotid plaque vulnerability classification.
|
||
|
561-02-005.
Respiratory Motion Classification and Outlier Rejection for Robust Motion Compensated Free-Breathing 1H Lung MRI
Impact: The new respiratory motion classification and outlier rejection algorithms can enhance the robustness of current state-of-the-art respiratory motion correction techniques and pave the road to the application of free-breathing ¹H lung MRI to patients with irregular breathing patterns.
|
||
|
561-02-006.
Diagnosis of Primary chyluria Based on MR Lymphangiography: A Comparative Study with CT Lymphangiography
Impact: MR lymphangiography can diagnose primary chyluria and is correlated with the results of CT lymphangiography, providing a comprehensive framework for disease classification, individualized treatment, and non-invasive follow-up.
|
||
|
561-02-007.
Impact of MR task aware data pre-processing and feature collation for robust Foundation model-based performance
Impact: This study provides a critical blueprint for leveraging foundation models in medical imaging. By tailoring data pre-processing & feature collation to specific clinical tasks, our work enables more robust and accurate AI, paving the way for reliable automation in diagnostic workflows.
|
||
|
561-02-008.
Automated Image Quality Evaluation of Cine Cardiac MRI Using a Convolutional Neural Network
Impact: Utilizing the CNN model, a prototype application was developed for automated, in-line detection of poor-quality cine images,
rescan of only affected slices, and replacement of poor-quality slices
with the rescanned slices, which could improve the efficiency and reliability
for CMR.
|
||
|
561-02-009.
Automatic MRI contrast classification combining metadata and interpretable image-based features
Impact: Accurate and interpretable automatic contrast classification enables robust large-scale MRI analyses by reducing human error and ensuring consistent labeling. This supports reproducible neuroimaging research and facilitates reliable image processing pipelines, ultimately accelerating both clinical and research workflows.
|
||
|
561-02-010.
Radiomics-Based Clustering Reveals Distinct Phenotypic Subgroups of Classical Menière’s disease in MRI: A Multicenter Study
Impact: This study identified five MD subgroups with distinct clinical and imaging profiles. This classification framework provides an objective foundation for precise medicine. The methodological approach may serve as a template for investigating heterogeneity in other disorders.
|
||
|
561-02-011.
Correlation Between Left Heart Myocardial Strain Based on CMR-FT and Heart Failure Classification in DCM
Impact: Myocardial strain technology can aid clinicians in precisely grading heart failure and developing more suitable treatment plans for patients with dilated cardiomyopathy.
|
||
|
561-02-012.
Adaptive ROI Extraction and 3D Consistency Voting for Robust Automated Water–Fat Classification in Musculoskeletal MRI
Impact: This framework enables reliable automated water–fat classification across complex musculoskeletal regions, supporting accurate assessment of inflammatory and degenerative disorders, reducing manual correction needs, and facilitating large-scale, multi-regional quantitative studies with improved robustness and reproducibility.
|
||
|
561-02-013.
Relaxometry-Based Classification of Human Knee Cartilage Degeneration
Impact: Accurate classification and prediction of OA severity
using clinically viable qMRI can promote faster and more accurate diagnosis of
early- to mid-onset OA, enabling earlier adoption of preventative treatment
options.
|
||
|
561-02-014.
Automatic Fibroglandular Tissue Volume Estimation for Workflow Efficiency in Breast Cancer Screening
Impact: Dense breast can increase
the risk of breast cancer. MRI plays a
crucial role and is feasible with Abbreviated Protocols. FGT Volume based triaging will improve the screening workflow efficiency and an indicator for follow up scans and preventive measures.
|
||
|
561-02-015.
Machine Learning with Multiparametric MRI and Clinical Biomarkers for Noninvasive Renal Interstitial Fibrosis Staging
Impact: This integrated machine learning approach provides an accurate, non-invasive tool for RIF staging across diverse kidney diseases, potentially guiding treatment decisions and serving as a biopsy alternative for high-risk patients, thereby enhancing clinical management.
|
||
|
561-02-016.
Analysis of MRI Prostate Volume Quantification: Ellipsoid vs 3D Modeling Methods
Impact: Accurate determination of prostate volume is important for diagnosis and treatment of various pathologies. This study provides a pathway for implementing ellipse approximation or 3D modeling methods with confidence and reliability. 3D printed models offer a quality assurance tool.
|