ReMemory

Inspiration

One of our team member’s grandmother suffers from dementia and encounters challenges in remembering the important people in her daily life. We were inspired to help find an effective solution to a disease that affects nearly 50 million individuals across the world. If our app can help patients with dementia remember a single member of their family and assist in their cognitive strength, then we have succeeded.

What it does

Users will start at the dashboard where they can select pictures of important people in their daily lives. We’ve also included a facial detection model on our Dashboard page that will determine who’s in the picture. Next is the contact page where users can create a contact anytime by adding a picture, name, relation, and a unique, important memory you have of them. This data will be used in testing for cognitive strength through randomized quizzes. You can start a quiz and review past quizzes through the dashboard page. Finally, you can see analytics on the quizzes which will help the user see which people they remember more versus less.

How we built it

We utilized Flutter with a firebase backend to personalize the mobile application to have a smooth user interface and authorization system. Additionally, we built a facial detection model using Firebase ML Kit and Teachable Machine. This will determine if the face from a picture you uploaded matches with a picture from the contact list. We also had an algorithm that randomized quizzes with unique questions every time. This was done by indexing all the contact info for each user in a list and using a random number generator to select a contact and their info for each question.

Challenges we ran into

We had a lot of difficulty running the facial detection machine learning model from Firebase ML kit on the IOS and Android emulators. After further research, we realized the problem was that we needed to download certain plugins for our Android and IOS workspaces which would allow them to connect to the Firebase backend. We also had problems randomizing the quizzes because there is a lot of information for each contact such as the relation and

Accomplishments that we're proud of

As a team, we’re most proud of our app’s facial detection machine learning model and the randomized quizzes. Both these were also some of our app’s greatest challenges, but we overcame them. The facial detection machine learning was

What we learned

In the process of developing this application, we learned how to integrate machine learning libraries into the Flutter User Interface. For example, facial detection with Firebase ML Kit required downloading a lot of packages and setting up the iOS and Android environments. We also learned how to save images to a storage bucket in Firebase.

What's next for ReMemory

In the future, we hope to continue developing ReMemory to create new features to assist patients with dementia to stimulate their minds to improve their cognitive abilities. This will take additional scientific research, but more importantly, the accuracy and ability of our tests to help patients will need to be ensured. For this to be achieved, we will require extensive alpha and beta testing before it’s put into public use.