Lymely

Your guide to staying ahead of Lyme disease.

View Demo  |  Slides  |  Poster

[UPDATED 04/07/2025]: Lymely now supports local inference with MLX for improved user security and affordability.

Table of Contents

1. About The Project
2. Features
3. Usage
4. Setup
5. Codebase Overview
6. Attribution

About The Project

Lymely is an entry for the 2024 Congressional App Challenge, designed to provide users with tools and resources to manage and understand Lyme disease risks.

(back to top)

Features

• Symptom and Activity Reporting with History 📋

  • Track symptoms and activities with a detailed history view.

• Interactive Tick Territory Heat Map 🗺️

  • Visualize tick territories with an interactive heat map.

• Lyme Education Resources 📚

  • Access a wealth of resources to learn more about Lyme disease.

• Rash Analysis Computer Vision Model 🔍

  • Analyze rashes using a computer vision model to assess potential Lyme disease risk.

• Risk Assessment with GPT-4o Agent ⚠️

  • Utilize a fine-tuned GPT-4o agent to assess Lyme disease risk.

(back to top)

Usage

Lymely is not yet available on the App Store, but you can still access the app by downloading the source code and running it on your own device.

(back to top)

Setup

To run Lymely on your local machine, you need to perform the following steps:

1. Clone the repository to your local machine.
2. Open the project in Xcode.
3. Create an Info.plist file in the Lymely/Lymely directory if it doesn't already exist.
4. Add the following keys to your Info.plist file:
   • OpenAIAPIKey: Your OpenAI API key
   • MBXAccessToken: Your Mapbox access token
5. Ensure you have the following libraries installed and imported in your project:
   • SwiftUI
   • UIKit
   • MapboxMaps
   • CoreLocation
   • Foundation
   • CoreML
   • Vision
   • MapKit
   • MapboxCoreMaps
   • MapboxCommon
   • AVFoundation
   • Combine

You can add these libraries to your project using Swift Package Manager or CocoaPods, depending on your preference.

(back to top)

Codebase Overview

• Lymely/Lymely.xcodeproj/project.pbxproj: The main project file for Xcode, containing the configuration for building the app.

  startLine: 1
  endLine: 626
  

• Lymely/Lymely/ContentView.swift: The main view of the app, providing access to various features including rash classification and survey taking.

  startLine: 1
  endLine: 239

• Lymely/Lymely/SurveyView.swift: Manages the survey data collection and risk assessment report generation.

  startLine: 1
  endLine: 687

• Lymely/Lymely/Resources/Map/script.py: A script for processing geographical data related to Lyme disease.

  startLine: 1
  endLine: 20

• Lymely/Lymely.xcodeproj/project.xcworkspace/contents.xcworkspacedata: Contains workspace data for the Xcode project.

  startLine: 1
  endLine: 8
  

• Lymely/LymelyUITests/LymelyUITests.swift: Contains UI tests for the Lymely app.

  startLine: 1
  endLine: 32

• Lymely/LymelyTests/LymelyTests.swift: Contains unit tests for the Lymely app.

  startLine: 1
  endLine: 17

(back to top)

Acknowledgments

• Hopkins Lyme Tracker - The data used to create the tick map is from Hopkins' Lyme Tracker.
• CDC Lyme Disease - The early Lyme disease symptoms used to educate the Llama 3.2 model are from the CDC.
• Lyme Disease Erythema Migrans Rashes Dataset - The dataset used to train the rash classification model, courtesy of Edward Zhang at Dartmouth College.

Attribution

@misc{lymely2024,
  author = {Elijah Renner},
  title = {Lymely: An App for Lyme Disease Awareness and 
  Management},
  year = {2024},
  howpublished = {\url{https://github.com/elijahrenner/
  lymely}},
}

(back to top)