TheBackPak

Inspiration

TheBackPak was inspired by the need for a practical, accessible solution to address posture issues in daily life. Poor posture is a common problem that can lead to chronic pain, reduced productivity, and long-term health issues. With advancements in wearable technology and data analytics, we saw an opportunity to create a tool that empowers users to monitor and improve their posture seamlessly. Our aim was to blend technology and health to promote better habits and long-term wellness.

What it does

TheBackPak is a wearable system that tracks the angle of the user’s back in real-time sensors. These sensors are strategically attached to clothing to monitor posture. The data collected by the sensors is processed through a linear regression model that differentiates between good and bad posture. The results are sent to a web app, which provides live feedback and alerts users when they need to adjust their posture. This ensures that users receive immediate, actionable insights for maintaining healthy posture habits.

How we built it

We started by integrating two sensors with a microcomputer. The sensors were configured with unique I2C addresses to allow simultaneous data collection. Python scripts were developed to read the x, y, and z data from the sensors, which was then processed using a linear regression model trained to classify posture as good or bad. Flask was used to create the web interface, which receives the processed data and updates the user interface in real-time. The entire system runs on the microcomputer, allowing seamless communication between the hardware and the web app.

Challenges we ran into

One of the main challenges was configuring the I2C addresses of both sensors to ensure they could operate simultaneously without interference. Additionally, ensuring accurate data collection from the sensors and aligning it with the linear regression model required careful calibration and testing. Integrating the web app with real-time data updates was another hurdle, as we needed to ensure smooth communication between Flask and the Python scripts handling the sensor data.

Accomplishments that we're proud of

We’re proud of building a fully functional prototype that successfully tracks posture and provides real-time feedback. Achieving seamless integration between hardware and software, as well as creating a user-friendly web interface, were significant milestones. We’re especially proud of developing a solution that has the potential to make a positive impact on users' health and well-being.

What we learned

Throughout this project, we deepened our understanding of working with sensor data and implementing machine learning models for practical applications. We also gained valuable experience in building end-to-end systems that require coordination between hardware components and software platforms. Troubleshooting real-time data transmission and enhancing the accuracy of machine learning predictions were key learning points.

What's next for TheBackPak

The next steps for TheBackPak include refining the model to improve accuracy and reliability and exploring additional features such as long-term data tracking and posture improvement suggestions. We plan to enhance the web app to include personalized user settings and analytics to offer insights over time. Additionally, integrating mobile app support and expanding compatibility with more types of clothing will help broaden the user base and utility of TheBackPak.