PolyPhone

With some of our members' knowledge of the biotechnology field we understood the need for quantity and quality data. From engineers building prosthetics and rehabilitation machines to sports scientists and physicians monitoring the post-injury recovery of an athlete they all needed data to achieve their goals, specifically gait data (locomotion achieved through the movement of human limbs). Currently, the most common way to analyse a person's gait is through a treadmill and a high-tech, expensive set of cameras. Not only is this inaccessible to a large number of people that would like to benefit from gait analysis but it also does not provide a method for collection of data in different environments (running on a beach, hiking a mountain etc.).

Our hack focuses on gathering gait data (angle of rotation of joints) using a magnetometer and then using the Firebase service we redirect the data in real-time to our Android app. Initially we were aiming to use accelerometer and gyroscope sensors attached to our thigh (for hip rotation), shin (for knee rotation) and foot (ankle rotation) with an Arduino and send the values via Bluetooth or Rasberry Pi server to our Android phone. However, we were unable to secure the required sensors and we are proud of coming up with the solution of using other Android phones to act as our sensors using their magnetometer (which measures angles compared to the Earths magnetic field). We did that by simply duct-taping our phones to the aforementioned body parts.

We built a 2-in-1 app that depending on the use we have for the particular phone we can have it acting as a transmitter (send data to Firebase) or as a receiver (receiving new data from Firebase as they change). The receiver can also store the information gathered offline on the device to be transferred to the user's database.

Coming into ICHACK19 none of us had any experience with building an Android application and having to extract the phone magnetometer's data in real-time was a real challenge for us and we are very proud to have solved this.

Going forward we would like to provide different options within the app. Most notably finding a way to create a 3D-render of the leg and (being really ambitious here) have another model with "perfect" form (e.g. that of a professional athlete) so that we can visually see the difference.