Related Experiment Video
Updated: May 5, 2026
09:17
Digital Spatial Profiling for Characterization of the Microenvironment in Adult-Type Diffusely Infiltrating Glioma
Published on: September 13, 2022
2.3K
MRI Deep Learning for Differentiating Glioblastoma, IDH Wild-type from Central Nervous System Diffuse Large B-cell
Mana Moassefi1, Paul A Decker2, Gian Marco Conte1
1Department of Radiology, Mayo Clinic, Rochester, Minnesota.
Cancer Research Communications
|May 4, 2026
Summary
Deep learning models using MRI scans can help distinguish between aggressive brain tumors like glioblastoma (GBM) and central nervous system diffuse large B-cell lymphoma (CNS-DLBCL). This non-invasive approach shows promise for accurate differential diagnosis in neuro-oncology.
Area of Science:
- Neuroscience
- Oncology
- Radiology
- Artificial Intelligence
Background:
- Glioblastoma, IDH-wildtype (GBM) and central nervous system diffuse large B-cell lymphoma (CNS-DLBCL) are aggressive brain tumors.
- These tumors often present with similar features on MRI, complicating diagnosis and treatment.
- There is a need for non-invasive methods to differentiate between GBM and CNS-DLBCL.
Purpose of the Study:
- To develop and validate deep learning models for the non-invasive differentiation of GBM and CNS-DLBCL using MRI data.
- To compare the performance of two distinct deep learning approaches (ensemble and loss minimization) for this classification task.
Main Methods:
- A three-stage temporal study design was employed, including model development on 146 CNS-DLBCL and 146 matched GBM cases.
- Models were trained using T1 post-contrast and T2-weighted MRI sequences.
- Two analytical approaches, an ensemble method and a loss minimization method, were compared on independent test cohorts, including prospective and external data.
Main Results:
- The deep learning models achieved an AUC of 0.84 (ensemble) and 0.83 (loss approach) on the prospective test cohort.
- Model performance remained consistent across different age groups and sexes.
- The ensemble model's prediction stability improved with an increased number of contributing models.
Conclusions:
- Deep learning models applied to MRI data are feasible for differentiating GBM from CNS-DLBCL.
- This AI-driven approach offers a promising non-invasive tool to aid in the differential diagnosis of these challenging brain tumors.
- The study highlights the potential of AI in neuro-oncology for improving diagnostic accuracy and guiding treatment decisions.
