Precovid

The website was created to support citizens across Latin America in self-assessing potential COVID-19 symptoms and accessing up-to-date information and personalized guidance based on their health status. I developed this project during my third semester of Commercial Engineering, while also participating in Startup School—an entrepreneurship and innovation track at the School of Engineering, Adolfo Ibáñez University.

Our team includes doctors, engineers, and journalists. Although most of us had never met in person, we came together with a shared purpose: to save lives.

GEOEDUCA: Javiera Gonzalez, Valentina Cáceres, Francisco Concha, Francisco Vergara. CONMAPAS: Jorge Ulloa, Salvador Muñoz, Gino Escobar. AWAKELABS: Gustavo Rodriguez, Ricardo Hermosilla. UAI STARTUP SCHOOL: Alfredo Osorio, Matías Donoso, Raimundo Valdivia. OTROS COLABORADORES: Claudio Moreno, Alberto Roa, Iván González, Matías Lobos, Paula Mardones, Carla Mardones, Leonardo Ramos, Matías Quinteros, Kurt Müller, Dino Cisterna, Daniela Bustos, Javiera Rosales, Evelyn Neira, Cristian Gonzalez, Daniela Bustos, Emiliano Araneda...

Table of Contents

• Description
• Requirements
• Features
• Quick Start
  • Requirements
  • Run locally
  • Directory Structure
• Directory Structure
• How It Works
  • User Flow
  • COVID-19 Symptom Detection Logic
    • Primary Symptoms
    • Contact History
    • Secondary Symptoms
    • Result Determination
    • Differential Diagnosis
  • Technical Details
• Privacy & Data Handling
• Developer Notes
• Troubleshooting
• License

Requirements

• PHP 7.2 or higher
• PHP extensions: curl, json, mbstring
• Web server: Apache, Nginx, or PHP built-in server
• See requirements.txt for details

Features

• COVID-19 self-assessment
• Personalized recommendations
• No persistent user data
• Responsive design
• Easy to deploy (PHP built-in server or any web server)
• Minimal dependencies

Quick Start

1. Requirements:

   • PHP 7.2 or higher
   • A web server (e.g., Apache, Nginx) or PHP built-in server

2. Run locally:

   cd app
   php -S localhost:8000 router.php

   Then open http://localhost:8000 in your browser.

3. Directory Structure:

   • app/ - Main application code
     • index.php - Main entry point and controller
     • router.php - Handles routing and static file serving
     • views/ - All user-facing pages (login, form, results, privacy, etc.)
     • static/ - Static resources (CSS, JS, images, JSON data)
   • panel/ - Admin or panel area (minimal)
   • docs/ - Documentation, presentations, logos, and media
   • README.md - This file

Directory Structure

app/
  index.php
  router.php
  views/
  static/
    css/
    js/
    images/
    json/
docs/
  DEVELOPMENT.md
  images/
panel/
README.md

How It Works

1. User Flow:

   • The user lands on the home page and is prompted to log in with basic information (name, country, email, etc.).
   • After accepting terms, the user proceeds to a symptom self-assessment form.
   • The form asks about key COVID-19 symptoms (shortness of breath, fever, cough, etc.).
   • Upon submission, the app calculates a risk score and redirects the user to a result page (positive/negative) with recommendations.
   • The user can share their result via WhatsApp and is encouraged to re-evaluate after 12 hours if symptoms persist.

2. COVID-19 Symptom Detection Logic: The application uses a sophisticated weighted scoring system based on early COVID-19 research and clinical observations. The detection process involves:

   a) Primary Symptoms (20 points each):

   • Shortness of breath (key respiratory symptom)
   • Fever (common in COVID-19 cases)
   • Cough (persistent dry cough is typical)

   b) Contact History (30 points):

   • Contact with a confirmed COVID-19 case

   c) Secondary Symptoms (12.5 points each):

   • Nasal mucus (less common in COVID-19)
   • Muscle pain (common but not specific)
   • Gastrointestinal symptoms (less common in COVID-19)
   • Duration of symptoms (longer symptoms may indicate COVID-19)

   d) Result Determination: A positive result is triggered if either:

   • The total probability score is ≥ 50% (multiple symptoms present)
   • All three primary symptoms are present (shortness of breath, fever, cough)

   e) Differential Diagnosis:

   • Some symptoms have negative weights when absent to reduce false positives
   • This helps differentiate COVID-19 from other respiratory conditions
   • The system prioritizes identifying potential cases while minimizing false negatives

3. Technical Details:

   • The main logic is in app/index.php.
   • Form data is sanitized and sent to an external endpoint (precovid.conmapas.cl/nuevaTest) for logging.
   • The risk score is calculated based on symptom answers (see code for details).
   • Static data (countries, regions) is loaded from JSON files in app/static/json/.
   • The app uses Bootstrap for styling and includes Google Analytics and Facebook Pixel for analytics.

Privacy & Data Handling

• The application does not provide medical diagnosis or emergency services.
• All data is collected for public health interest and is not used for commercial purposes.
• Users must accept the privacy policy and terms before using the service.
• See the Privacy Policy for full details.

Developer Notes

• Main files:
  • index.php: Handles routing, form processing, and result logic.
  • views/: Contains all HTML/PHP templates for the user interface.
  • static/: Static assets (CSS, JS, images, JSON data).
• Adding questions: Edit views/formulario.php to add or modify symptom questions.
• Updating regions/countries: Edit the relevant JSON files in static/json/.
• No persistent database is used; data is sent to an external API.
• Contributions: PRs and issues are welcome! Please document any new features or changes.

Troubleshooting

• Address already in use: Stop other PHP servers or use a different port (e.g., php -S localhost:8080 router.php).
• Static files not loading: Ensure you are using the correct /static/ paths and that the static/ folder exists with the right subfolders.
• Permission errors: Make sure your user has read access to all files in the project directory.

License

This project is for educational and public health purposes. For reuse or adaptation, please contact the original authors or open an issue.