Brain Tumor Monitoring System Documentation

Welcome to the comprehensive documentation for the Brain Tumor Monitoring System, a sophisticated MLOps solution for monitoring and detecting data drift in brain tumor image classification models.

Overview

The Brain Tumor Monitoring System is designed to provide comprehensive monitoring capabilities for machine learning models that classify brain tumor images. The system includes:

Real-time Monitoring: Continuous monitoring of model predictions and data quality
Drift Detection: Advanced algorithms to detect data drift in image features
Automated Reporting: HTML reports with visualizations using Evidently AI
RESTful API: FastAPI-based backend with comprehensive endpoints
Modern Frontend: React-based dashboard with real-time updates
Cloud Deployment: Ready for deployment on GCP, AWS, or Azure

Key Features

Image Feature Extraction: Comprehensive feature extraction from brain tumor images
Statistical Analysis: Mean, standard deviation, entropy, and other statistical measures
Drift Detection: Customizable thresholds for detecting significant data drift
Dashboard: Real-time monitoring dashboard with key metrics
API Integration: Seamless integration with existing ML pipelines
Scalable Architecture: Designed for production deployment

Quick Start

# Clone the repository
git clone <repository-url>
cd brain-tumor-monitoring

# Install dependencies
pip install -r requirements.txt

# Set up database
export DATABASE_URL="postgresql://user:password@localhost:5432/monitoring"

# Run the backend
uvicorn backend.src.api:app --reload

# Run the frontend
cd frontend && npm install && npm start

For detailed installation instructions, see Installation Guide.

API Reference

The system provides a comprehensive REST API for monitoring operations:

Health Checks: /health
Monitoring Dashboard: /monitoring/dashboard
Drift Reports: /monitoring/drift-report
Feature Analysis: /monitoring/feature-analysis
Data Quality: /monitoring/data-quality

For complete API documentation, see api/index.

Monitoring System

The monitoring system provides:

Feature Extraction: Automatic extraction of image features
Drift Detection: Statistical analysis for data drift
Reporting: HTML reports with visualizations
Alerting: Configurable alerts for drift detection

For detailed monitoring documentation, see monitoring/index.

Machine Learning

The ML pipeline includes:

Model Training: YOLOv8-based tumor detection
Prediction Pipeline: Real-time image classification
Feature Engineering: Comprehensive feature extraction
Model Versioning: Version control for ML models

For ML documentation, see ml/index.

Frontend Dashboard

The React-based frontend provides:

Real-time Monitoring: Live updates of system metrics
Interactive Charts: Visual representation of drift data
Report Viewer: HTML report display
Responsive Design: Mobile-friendly interface

For frontend documentation, see frontend/index.

Contributing

We welcome contributions! Please see contributing for guidelines.