About a decade ago, my grandma started getting dementia. At first, she would forget details about me and other family members, but then it started to get worse--to the point where she would forget my name. My family felt a lot of pain in being forgotten by someone we truly loved. However, in reality, my grandma was the one going through the most. I wanted to build an app that would help alleviate this stress.

What it does

With a computer and webcam set up in a patients room, this app uses facial recognition to identify loved ones from a database of faces. It uses text-to-speech to announce to the patient who has arrived in the room. It gives their name, their relation, and a quick blurb about who they are

How we built it

We built it using Flask, javascript, face_recogniton in python, and Amazon Polly.

Challenges we ran into

Our main issues we faced were in data transfer and storage of the faces to our online database. It had to capture and encode the faces and efficiently store them for use in later recognition.

Accomplishments that we're proud of

We're very proud we got this to work, especially with Amazon Polly.

What we learned

What's next for Dementia Helper

We want to further improve upon the mechanism for primary caregivers to input meaningful information for the patients. We want to build a portal to give the caregivers full access to helping the patients.

Other features: An alert that says when a patient hasn't been visited in awhile. An end of the day summary of visits

Built With

Share this project: