DeepBraTumIA
DeepBraTumIA is a deep learning-based software tool for automated segmentation of glioblastoma multiforme (GBM) multisequence MR images. It enables automated skull stripping and voxel-wise segmentation of the brain tumor and its main sub-compartments (necrotic, contrast-enhancing, and edema).
Input Requirements
The software requires the following MRI sequences:
T1-weighted, T2-weighted, T1-contrast, FLAIR.
Key Features:
- Automated Segmentation: Full segmentation of tumor compartments with volumetric measurements.
- Batch Processing: "Batch mode" enabled for processing multiple cases efficiently.
- Resource Aware: Automatically analyzes hardware resources and adapts memory and CPU/GPU requirements.
- Versatility: Usable on both pre- and post-operative cases (optimal performance on pre-operative).
To request software access, please use the "Download DeepBraTumIA" link on the left menu).
Clinical Application
For patient care and clinical workflows:
Our research has been translated into FDA-cleared software solutions via Neosoma Inc. Clinicians seeking a certified tool for daily practice should visit our commercial partner.
Visit Neosoma Inc.
Research & Development
For academic research and code inspection:
Researchers interested in the underlying algorithms, reproducibility, or open-source collaboration can access our code repositories.
Visit MIA Lab Repository
