Patient Main Screen
Patient People Screen
Patient People Profile Popup
We have always wanted to work on an application to help Alzheimer's patient or people suffering from dementia; although none of us have ever experienced it so we've always been scared to work on it during previous hackathons. This time we wanted to take on the challenge, so we researched a lot on the topic and were able to come up with CareBuddy.
What it does
CareBuddy is an elderly friendly application that helps patients suffering from any form of dementia or Alzheimer's to remember past events and record new ones. Our app enables a communication between the caregiver and the patient, therefore it has two sides to it:
- The patient's side: Enables some features like facial recognition, which the patient can use to take a picture of the people they do not recognize and get an idea of who they are. in addition there are some announcements given my the caretaker and a chat option.
- The caregiver's side: the caregiver acts as an admin to our application. they hold most of the write rights. They add the patients family member and personal informations. the caregiver can also post reminders that will show on the patient's side as well as discuss with the patient through our chat.
How we built it
We used react native to make the mobile application. For the facial recognition, we used the Google Vision API. After taking the picture, we send a base64 encoded version of the picture to the api and we use the results to draw a rectangle around the picture and output a text letting the patient know who he just took a picture of.
Challenges we ran into
We all had all sorts of issues with react native considering the fact that we are fairly new to it. Another challenge was getting the rectangle to show over the picture. we had to deal with image resizing in order to get it to work. In order to show the identity of the person in the picture we actually had to use some hard-coded description. I we had more time we would have been able to train an AI to recognize the patients family member and if a person is not recognized, the caregiver should receive a request to add the new person if they are related to the patient.
Accomplishments that we're proud of
Being able to get the image recognition to work is our biggest accomplishment along with finishing most part of our hack for the demo
What we learned
We learned more about some components from expo like Camera, ImageBackground and other
What's next for CareBuddy
we would like to continue working on it and implement the functionalities we were not able to finish such as the chat and an "add reminder" option on the care giver's side **