Choose the type of MRI data you have.
A single MRI image can be screened for four currently supported classes. A complete multi-sequence MRI study can additionally undergo detailed three-dimensional tumor-region segmentation.
A single MRI image can be screened for four currently supported classes. A complete multi-sequence MRI study can additionally undergo detailed three-dimensional tumor-region segmentation.
Single MRI Image
![MRI preview]()
Tumor Detection & Classification
Upload one brain MRI image in JPG, JPEG, PNG or BMP format.
The current AI classifier predicts:
- No Tumor
- Glioma
- Meningioma
- Pituitary Tumor
- Neurocytoma
- Schwannoma
- Other / Uncertain
Class probabilities
This pathway performs classification from a single
2D MRI image. It does not provide 3D tumor
segmentation. The output is an AI imaging prediction
and not a definitive clinical diagnosis.
Complete 3D MRI Study
Open Full MRI Study Analysis
Detailed Tumor Segmentation
Use this pathway when the complete MRI study contains:
- T1
- T1 contrast-enhanced (T1ce)
- T2
- FLAIR
The MRI Deployment Model produces:
- Whole Tumor segmentation
- Tumor Core segmentation
- Enhancing Tumor segmentation
- Tumor volume measurements
- Interactive 3D MRI visualization
- Downloadable masks
- PDF analysis report
Open Full MRI Study Analysis
This pathway currently requires all four supported
MRI sequences. It is designed primarily for detailed
glioma-region segmentation.
Current Single-Image Classifier Performance
Results below are from the held-out 1,600-image test set.
95.76%
Test Accuracy
95.84%
Macro F1
100%
No-Tumor Recall
100%
Pituitary Recall
99.0%
Meningioma Recall
82.75%
Glioma Recall
These values describe performance on the held-out research
dataset. They should not be interpreted as guaranteed
performance on arbitrary hospital images.
