neurodegenerAI

Neuro-Trends Suite

Integrated Platform for Neurodegenerative Analysis & Social Intelligence

A consolidated, hardened system combining advanced neuroscience modeling with real-time social media trend detection. Built for medical research, clinical decision support, and market intelligence.

Architecture Overview

The suite has been completely refactored from a fragmented prototype into a unified, production-grade stack:

Core Components

Core API (core_api/): Unified FastAPI gateway serving both Neuro and Trends endpoints
Unified Dashboard (dashboard_app.py): Premium Streamlit interface for all analysis modes
PostgreSQL Backend: Mandatory persistence layer for all predictions and social data
Shared Libraries (shared/lib/): Common utilities including PII scrubbing, logging, and database management

Legacy Components (Deprecated)

The following directories are legacy and will be removed in future versions:

neurodegenerai/src/api/ -> Migrated to core_api/src/api/v1/endpoints/neuro.py
trend-detector/src/api/ -> Migrated to core_api/src/api/v1/endpoints/trends.py

Quick Start

Prerequisites

Docker & Docker Compose
Python 3.11+ (for local development)

Launch the Full Stack

# Start all services (API, Dashboard, PostgreSQL, pgAdmin)
docker-compose up --build

Local Development

# Install dependencies
pip install -r requirements.txt

# Run Core API
export PYTHONPATH=$PYTHONPATH:.
python -m uvicorn core_api.src.api.main:app --host 0.0.0.0 --port 8000

# Run Dashboard (in separate terminal)
streamlit run dashboard_app.py --server.port 8501

Features

NeuroDegenerAI

Tabular Biomarker Analysis: Multi-modal ensemble predictions using LightGBM/XGBoost
MRI Structural Analysis: 3D CNN for brain imaging classification
EEG Time-Series Decoding: Real-time neurological state detection
Medical Data Validation: Enforced schemas for ADNI-compliant data ingestion
PII Protection: Automatic scrubbing of sensitive patient information

Trend Detector

Real-Time Social Monitoring: Reddit/Twitter stream ingestion with exponential backoff
Topic Clustering: BERTopic-powered semantic analysis
Trend Intelligence: Growth rate tracking and anomaly detection
Semantic Search: Vector-based social data retrieval

Security & Privacy

Implemented Safeguards

PII Scrubber: Automatic redaction of emails, SSNs, phone numbers, and DOBs
Data Validation: Medical-grade range checks for all biomarker inputs
Structured Logging: Service-tagged JSON logs with no sensitive data leakage
Database Isolation: PostgreSQL with connection pooling and prepared statements

Environment Configuration

Create a .env file:

# Database
POSTGRES_USER=trends
POSTGRES_PASSWORD=your_secure_password
POSTGRES_DB=trends

# Reddit API (Optional)
REDDIT_CLIENT_ID=your_client_id
REDDIT_CLIENT_SECRET=your_client_secret
REDDIT_USER_AGENT=neuro-trends-suite/1.0

# System
ENV=production
LOG_LEVEL=INFO
NEURO_DEMO_MODE=false

API Endpoints

Neuro Analysis

POST /v1/neuro/tabular - Biomarker prediction
POST /v1/neuro/mri - MRI volume analysis
POST /v1/neuro/eeg - EEG state decoding

Trend Detection

GET /v1/trends/top - Top trending topics
POST /v1/trends/search - Semantic search

System

GET /health - Health check with DB status

Full API documentation: http://127.0.0.1:8000/docs

Testing & Validation

# Run all tests
pytest

# Verify Core API
curl http://127.0.0.1:8000/health

# Test PII Scrubber
python -c "from shared.lib.io_utils import PIIScrubber; s = PIIScrubber(); print(s.scrub('Contact: john@example.com'))"

Database Schema

Tables

patients - Patient demographics (anonymized)
neuro_predictions - All model predictions with metadata
social_posts - Ingested social media data
trend_topics - Detected trending topics with scores

Migrations

# Auto-create tables on first run
docker-compose up

# Manual migration (if needed)
alembic upgrade head

Troubleshooting

Common Issues

API won’t start:

# Check PostgreSQL is running
docker-compose ps

# View logs
docker-compose logs core-api

Dashboard connection error:

# Ensure Core API is accessible
curl http://127.0.0.1:8000/health

# Check CORE_API_URL in dashboard_app.py

Import errors:

# Verify PYTHONPATH
export PYTHONPATH=$PYTHONPATH:.

License

MIT License - See LICENSE file for details

Medical Disclaimer

This software is for research and educational purposes only. It is not FDA-approved and should not be used for clinical diagnosis without proper validation and regulatory approval.

Built with: FastAPI, Streamlit, PostgreSQL, PyTorch, Scikit-learn, BERTopic, PRAW

This site is open source. Improve this page.