Skip to content

nathadriele/neuroPredict-precision-medicine-system

Repository files navigation

NeuroPredict: Precision Medicine System for Refractory Epilepsy

Captura de tela de 2025-12-02 11-47-23

Python 3.10+

Overview

NeuroPredict is an AI-driven precision medicine system that integrates clinical, genomic, EEG, neuroimaging, and medical literature data to predict individual treatment response in refractory epilepsy. The platform combines knowledge graphs, deep learning models, and large language models (LLMs) to provide evidence-based, personalized therapeutic recommendations.

Key Features

  • Multimodal Analysis: Integration of clinical, genomic, EEG, and neuroimaging data
  • Medical Knowledge Graph: Semantic representation of relationships between genes, drugs, symptoms, and phenotypes
  • Ensemble Learning Prediction: XGBoost, LightGBM, CatBoost, and neural network models
  • Explainability: SHAP values and feature importance analysis
  • Interactive Web Interface: Dashboard for visualization and analysis
  • RAG with LLMs: Retrieval-augmented generation system for evidence-based recommendations
  • MLOps Pipeline: Model versioning, monitoring, and CI/CD

Architecture

┌─────────────────────────────────────────────────────────────┐
│                     Camada de Ingestão                      │
│  (CSV, FHIR, DICOM, VCF, Literatura PubMed)                │
└────────────────────┬────────────────────────────────────────┘
                     │
┌────────────────────▼────────────────────────────────────────┐
│                  Camada de Processamento                     │
│  • ETL Pipeline          • Feature Engineering              │
│  • Data Validation       • Normalização                     │
└────────────────────┬────────────────────────────────────────┘
                     │
         ┌───────────┴───────────┐
         │                       │
┌────────▼────────┐    ┌────────▼─────────┐
│  Knowledge Graph │    │  Feature Store   │
│   (Neo4j)        │    │   (Feast)        │
└────────┬────────┘    └────────┬─────────┘
         │                       │
         └───────────┬───────────┘
                     │
┌────────────────────▼────────────────────────────────────────┐
│                    Camada de ML/AI                          │
│  • Ensemble Models    • GNN (PyG)      • Transformers       │
│  • SHAP Explainer     • AutoML         • RAG System         │
└────────────────────┬────────────────────────────────────────┘
                     │
┌────────────────────▼────────────────────────────────────────┐
│                  Camada de Aplicação                        │
│  • REST API (FastAPI)  • Web Dashboard  • Relatórios        │
└─────────────────────────────────────────────────────────────┘

Prerequisites

  • Python 3.10+
  • Docker & Docker Compose
  • Neo4j 5.x
  • PostgreSQL 15+
  • At least 16GB RAM
  • GPU

Quick Installation

git clone https://github.com/nathadriele/neuroPredict-precision-medicine-system.git
cd neuropredict

python -m venv venv
source venv/bin/activate  # Linux/Mac
# or
venv\Scripts\activate  # Windows

pip install -e ".[dev]"

cp .env.example .env
docker-compose up -d

alembic upgrade head

python scripts/load_sample_data.py

Usage

1. Model Training

python -m neuropredict.train --config configs/training_config.yaml

python -m neuropredict.train --config configs/training_config.yaml --hpo

2. API REST

uvicorn neuropredict.api.main:app --reload --host 0.0.0.0 --port 8000

# http://localhost:8000/docs

3. Prediction

image

4. Web Dashboard

image

Testing

image

Performance

image

Contributing

Contributions are welcome:

  1. Fork the project
  2. Create a branch for your feature (git checkout -b feature/AmazingFeature)
  3. Commit your changes (git commit -m 'Add some AmazingFeature')
  4. Push to the branch (git push origin feature/AmazingFeature)
  5. Open a Pull Request

Important Notes

This system is for research and educational purposes only. It must not be used as a substitute for professional medical advice, diagnosis, or treatment. Always seek the advice of your physician or other qualified healthcare professional.

Quote

If you use this project in your research, cite:

@software{neuropredict2024,
  author = {Your Name},
  title = {NeuroPredict: Precision Medicine System for Refractory Epilepsy},
  year = {2024},
  url = {https://github.com/your-username/neuroPredict-precision-medicine-system}
}

About

NeuroPredict is a basic AI-based precision medicine project that integrates clinical, genomic, neuroimaging, and medical literature data to predict individual treatment response in refractory epilepsy. The system uses knowledge graphs, deep learning models, and LLMs to provide personalized therapeutic recommendations.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors