Model Ready
Research-Grade Deep Learning

Detect Brain Tumors
with Precision AI

Upload an MRI scan and our convolutional neural network instantly classifies it across four categories — Glioma, Meningioma, Pituitary, or No Tumor — with confidence scores.

4 Tumor Classes
CNN Architecture
128px Input Resolution
<1s Inference Time
🎲 Try a Sample Pick any class to load a random MRI from your dataset folder

Analyze Your MRI Scan

Supported formats: JPG, PNG, BMP, TIFF, WebP · Max size: 16 MB

Drop MRI Image Here

or

Awaiting Scan

Upload an MRI image to see AI analysis results here

Detectable Tumor Types

Our model is trained to classify MRI scans into the following categories

🔴

Glioma

High Risk

Originates from glial cells. The most common malignant primary brain tumor, representing ~33% of all brain tumors.

Most Common Variable Grade
🟠

Meningioma

Medium Risk

Arises from the meninges. Typically benign and slow-growing, though location can make treatment challenging.

Usually Benign Slow Growing
🟣

Pituitary Tumor

Medium Risk

Forms in the pituitary gland at the base of the brain. Can disrupt hormone regulation throughout the body.

Hormone Impact Often Treatable
🟢

No Tumor

Healthy

No pathological tumor growth detected in the scan. The MRI appears within normal neurological parameters.

Normal Scan Recommended Checkups

How It Works

Upload MRI Scan

Drop your brain MRI image in JPG, PNG, or similar format. The model accepts standard medical imaging exports.

CNN Processing

The image is resized to 128×128px, normalized, and passed through a 4-layer convolutional neural network trained on thousands of MRI scans.

Classification Results

Softmax probabilities are computed for all 4 classes. The highest-probability class is returned with confidence scores and clinical context.

Technology Stack

TensorFlow / Keras CNN Model
Python / Flask Backend API
OpenCV Image Processing
NumPy Array Operations
Render Cloud Deployment
Kaggle Dataset Training Data