ReCura: Comprehensive Health Monitoring Dashboard

Overview

ReCura is an integrated health monitoring system designed to provide real-time insights into patient health metrics and predict readmission risk. The platform uses machine learning models to analyze patient data and provide healthcare professionals and patients with actionable information.

Project Architecture

Application Workflow

Application Screenshots

Admin Dashboard	Patient Dashboard
Healthcare provider interface with analytics and model training capabilities	Patient interface showing real-time personalized health metrics and risk assessment

Architecture

The Project follows a three-tier architecture:

1. Data Preparation

• Dataset Search: Collection of patient health records from various sources
• Data Pipeline: Processing raw data into structured formats ready for analysis
• Random Data Generation: Temporary data creation for development and testing

2. Model Development

• Implementation of multiple prediction algorithms:
  • Logistic Regression
  • Random Forests
  • Decision Trees

3. Web Application Deployment

MongoDB (NoSQL Database)

• Cloud and local server options for data storage
• Real-time data collection and persistence
• Secure storage of patient records

Streamlit Web Applications

• Patient Dashboard: Personalized health metrics visualization
• Admin Dashboard: Healthcare provider interface with comprehensive analytics
• Authentication: Secure login system for both patients and administrators

Technology Stack

• Frontend: Streamlit for responsive web interface
• Backend: Python data processing and ML pipeline
• Database: MongoDB for flexible NoSQL storage
• Data Processing: Pandas, NumPy for efficient data manipulation
• Machine Learning: Scikit-learn for predictive modeling
• Visualization: Plotly, Matplotlib for interactive charts

Key Features

• Real-time health metrics monitoring
• Readmission risk prediction using multiple ML models
• Interactive data visualization
• Dual interface for patients and healthcare administrators
• Secure authentication system
• Cloud-based data storage

Environment Setup

ReCura requires the following environment variables:

• MONGODB_URI: Connection string for your MongoDB database

Create a .env file in the project root with these variables before running the application.

Recreate the MongoDB Atlas database

The original Atlas cluster used by this project no longer resolves and appears to have been deleted. Create a new Atlas deployment as follows:

1. Sign in at MongoDB Atlas and create a project.

2. Create a free M0 database deployment in a nearby region.

3. Under Database Access, create a dedicated database user with read/write access. Do not reuse your Atlas account password.

4. Under Network Access, add your current IP address for local development. For Streamlit Community Cloud, add the network access required by your deployment and use a strong, unique database password.

5. Click Connect → Drivers → Python, then copy the mongodb+srv:// URI.

6. From the project root, create the local configuration:

   cp .env.example .env

7. Replace the placeholders in .env. If the password contains reserved URL characters, URL-encode it. Never commit .env.

8. Test the connection and generate a sample record from either the project root or the worker directory:

   python mongodb_worker/mongodb_store_data.py

   The script now checks the Atlas connection before prompting for the interval and patient ID. Stop continuous generation with Ctrl+C.

For Streamlit Community Cloud, add MONGODB_URI in the app's Secrets settings instead of uploading .env.

Getting Started

1. Clone the repository
2. Install dependencies: pip install -r requirements.txt
3. Configure MongoDB connection
4. Run the application: streamlit run app.py
5. Run random data generation for a specific user ID: python mongodb_worker/mongodb_store_data.py

---

ReCura: One Health Dashboard - Proactive, precise, and perfectly tailored health management.