The motivation for this project was initially inspired by helping my grandma with her computer problems. I was helping her fix a relatively simple issue but she told me, to her, “it seemed like rocket science”. When my parents wanted her to get a smartwatch to help monitor her health, she was equally confused by technology. This idea got us thinking about the interaction between elderly people and technology and how it affects their daily lives. We ended up settling on trying to develop a tool that would simplify wearbile health data (ECG in our case) for the elderly population, since the data can be extremely useful but unfortunately not very easy to understand.

What it does

Our tool takes patients ECG readings from their wearable devices and uses a deep learning algorithm to detect if there is an anomaly in their heart readings. If the data is abnormal, our tool will send the important data to the patient's electronic medical records and provide them with a simple “normal” vs “abnormal” result. The user will also be displayed possible “next steps” such as the phone number for a local doctor (or family doctor) to possibly set up an appointment. Additional simple resources (such as videos) are then provided to the user so they can learn more about how to improve/maintain their cardio health and understand what their ECG signals mean.

How we built it

We built it by starting with a machine learning model that was improved upon (using TensorFlow) then created a simple web framework that would connect to our ECG algorithm.

Challenges we ran into

We had many challenges throughout the weekend ranging from the time zone difference (since almost every member of our team was in a different zone) as well as sleep deprivation. We also found obtaining useful large data sets to be more difficult than expected since some of the data sets we wanted to initially had labelling that was not clear or in a difficult format to understand.

Accomplishments that we are proud of

We are very proud of coming up with an idea and being able to execute it in such a short period of time and with very little experience. Moreover, we strongly believe that our project addresses the difficulties faced by the elderly regarding ECG wearables.

What we learned

From this project, we learned a ton about collaborating as a group virtually and how to successfully work on a project when all the members are spread across the world. During the current COVID times this an extremely useful skill and something we were happy to improve. We also learned a lot about developing learning algorithms and deploying a simple web app.Finally being that it was our first hackathon, we learned how to best approach a hackathon and work collaboratively. On top of that, we learned so much from the speakers and presenters.

What's next for Easy Beat

The next steps for Easy Beat would be making the connection between the website and the smartwatch, as well as enabling the submission of the patients’ ECG to their electronic medical record and making sure that all the patient data is protected.

Share this project: