Hybrid Deep Learning Framework for Brain Tumor Classification Using ConvNeXt and AttGRU

A hybrid deep learning framework utilizing a modified ConvNeXt architecture for feature extraction with SHAP-based feature selection and an Attention-based Gated Recurrent Unit (AttGRU) classifier for high-precision multiclass brain tumor classification from MRI images.

Analyze Medical Image

Experience cutting-edge AI-powered brain tumor classification. Upload your medical image for instant analysis using our advanced deep learning algorithm.

Upload Your Image

Drag & drop or click to select your medical image

Drop your image here

or click to browse from your device

JPG, PNG, BMP, TIFF
Max: 10MB

Try Sample Images

Test our AI with pre-loaded medical images for instant demo

Sample Dataset
5 Normal
5 Pituitary
5 Meningioma
5 Glioma

Model Performance Metrics

97.90%
Accuracy
Overall Correctness
90.40%
Sensitivity
True Positive Rate
98.57%
AUC Score
Area Under Curve
90.66%
Precision
Positive Predictive Value
90.29%
Specificity
True Negative Rate
87.95%
F1-Score
Harmonic Mean

How It Works

Our multi-stage deep learning pipeline ensures accurate and reliable brain tumor classification

1. Upload Image

Upload your medical image securely to our system

2. Preprocessing

Advanced image enhancement and normalization

3. AI Analysis

Deep learning model analyzes patterns and features

4. Results

Get detailed analysis report with confidence scores