Skip to main content

Architecture

Architecture

System Overview​

How everything works together:​

note

We do not have all this working functionality.

This is a high-level overview of what we were working towards and how the system would work together.

  1. The SmartSocks are the main hardware component.

    • It contains sensors that monitor foot temperature and pressure.
    • The data collected by the sensors is transmitted to a mobile application via Bluetooth.
    • The app analyzes the data and alerts the user if there are any signs of potential foot ulcers or any abnormalities.

  2. The mobile app is the main software component.

    • It receives and analyzes the data from the smart sock.
    • The app is responsible for alerting the user if there are any abnormalities with their readings.

  3. The web server backend is built with Flask.

    • It is responsible for storing and retrieving data from the MongoDB database.
    • The web server backend provides an API for the mobile app to communicate with the database.

  4. The MongoDB database stores the data collected by the smart sock.

    • The database is used to store historical data and provide insights into the user's foot health over time.
    • It is also used to retrieve data for the mobile app to display to the user via graphs, charts, etc.

  5. The user interacts with the mobile app to view real-time data on their foot health.

    • The app provides alerts and recommendations based on the data collected by the smart sock.
    • Notifications are sent to the user if there are any abnormalities detected in the readings.

  6. The user can also view historical data and trends in the mobile app.

    • This allows the user to track changes in their foot health over time and make informed decisions about their care.

Deep Dive into the components​

What we have so far:​

Smart Sock (Prototype - Raspberry Pi Pico W):​

  • We didn't have the time to actually build out the SmartSocks, but we did manage to create a prototype using a Raspberry Pi Pico W.
  • The prototype contains 8 temperature sensors and 8 pressure sensors (16 sensors total).
  • The Pico W sends a BLE advertisement with a GATT service that contains the sensor data in characteristic values.
    • Read more about Bluetooth Low Energy (BLE) here.

Mobile App​

  • The mobile application is built with Android Studios using Java.
  • The app is responsible for receiving and analyzing the data from the smart sock (Pico, in this case).
  • We were able to get a lot of functionality working, such as:
    • Scanning for BLE devices.
    • Connecting to the Pico.
    • Reading the sensor data from the Pico.
    • Displaying the sensor data in intervals.
    • Displaying the sensor data in a chart and heatmap.
    • Profile page
    • Debugging page
    • Registering and logging in users
  • For more information on the mobile app, check out the Android App section.

Web Server Backend​

  • The web server backend is built with Flask.
  • It is responsible for storing and retrieving data from the MongoDB database.
  • The web server backend provides an API for the mobile app to communicate with the database.
    • For more information on the web server and the endpoints available, check out the Flask Backend section.
  • The web server is also hosted on render and is accessible via the internet.
    • Please note that we can using the free tier of render, by default the server will go to sleep after 1 hour of inactivity. This means that the first request after the server has gone to sleep will take a few minutes/seconds to wake up the server.

MongoDB Database​

  • The MongoDB database stores the data collected by the Pico.
  • We have two collections in the database:
    • users - stores user information such as username, password and created date.
    • data - stores the sensor data collected by the Pico.
  • For a deeper dive into the database and the collections, check out the MongoDB section.

User Interface​

  • The user interacts with the mobile app to view their sensor data.
  • We have a bar chart and a heatmap that displays the sensor data upon request (doesn't update in real-time).
  • The user also has a Profile page where they can view and edit their information, such as: Name, Email, DOB, Weight, Height, Doctor's Name, and Doctor's Email.

Conclusion​

SafeStep is a very comprehensive full-stack project that involves a lot of moving parts.

We made a lot of progress in a short amount of time, especially considering that we were all very new to some the technologies we were using. SafeStep is a great example of how a team can come together and build something amazing in a short amount of time.

We hope that this project will inspire others to take on similar challenges and build something, no matter how big or small.