Innovape

The Juul add-on.
Our app.
Consumer statistics.

Inspiration

Juul has not only taken over the e-cigarette industry, but also people's lives. Nicotine content in one Juul pod is equivalent to those in a pack of 10 cigarettes. Addiction thus becomes a serious health problem, in particular to children and teens who obtain Juuls secondhand. Juul marketing plays to their vulnerability with flavors such as watermelon and strawberry lemonade. What can we make in a weekend that can reduce nicotine dependence and allow smokers to quit smoking by transitioning using Juul?

What it does

We created an attachment to a Juul that allows users to control the nicotine vape rate in real time through the Innovape mobile app. Users are provided the option to start a quitting plan, with individualized goal setting and projected outcomes given user behavior. Our app stays with the user throughout their journey and asks for their emotional state. Their feedback allows us to deliver a personalized nicotine reduction algorithm for each user, with the goal that the user will not notice the reduced nicotine over time. Innovape is compatible with any Juul device.

How we built it

We attached an Arduino to the Juul in order to control the vaporization rate. The Arduino communicates with the mobile app through an EC2 instance, where the Arduino sends data when the user inhales, and the mobile app sends signals to control the flow rate. We used the React-native framework for the app with rapid design iteration cycles made possible by Expo.

Challenges we ran into

The biggest challenge was that Juul's proprietary device prevented easy communication between Juul and a mobile device. Research on nicotine replacement therapy does not show promising quit rates. Though Juul collects many metrics from their users, those data are unavailable. The lack of data posed a significant challenge on our personalized nicotine reduction algorithm.

Accomplishments that we're proud of

We're very proud of our user-focused design of the product, such as the beautiful user interface on the Innovape app. We managed to work with unfamiliar voltages, overcame the proprietary design of the Juul hardware, and managed to integrate hardware with software in a tight timeline.

What we learned

How to program Arduinos, build integrated circuitry, how to communicate between a mobile device to unrelated hardware, and apply Gaussian statistical reasoning.

What's next for Innovape

Integrate the hardware add-on to an embedded contraption on the Juul itself. Add more heuristics to analyze user data and trends. With more data, the diverse user data can be used to make inference on user's diseases remotely, applying machine learning on the edge, allowing for real time disease detection.

Built With

Submitted to

Hack the North 2019
- Winner Hack the North 2019 Finalists

Created by

I worked on the hardware components (Juul + Arduino), mechanical design, and statistical analyses of usage data.

Winnie Xu
Computer Science and AI @ U of T
I worked on the statistical models that adjusts to each user's nicotine habits. I learned Gaussian Processes and how to distribute code with Flask.

Tianyu Lu
Computational Biology Student @ University of Toronto
I built the mobile app and connected it to the arduino via an EC2.

Hugo Burbelo
Reversed-engineered the juul, wrote the firmware, and did the electrical design.

Olivia Li