1 of 22

mHealth Hypertension

2 of 22

The Team

  • Yufan Xu - Frontend, GUI
  • Yunzhe Liu - Backend
  • Bo Fu - Frontend, GUI
  • Xingan Wan - Testing, Frontend

3 of 22

Technologies

EXPO-CLI

REACT-NATIVE

JAVASCRIPT

AWS

4 of 22

Backend Architecture

5 of 22

Application Walk-through

6 of 22

Home Page

  • Sign In
  • Sign Up

7 of 22

Sign Up

  • The user will received a confirmation code
  • The user is able to ask to resend a code
  • Proper frontend validation check is implemented

8 of 22

Dashboard

  • Progress
  • Check In
  • Daily Quests
  • Achievements
  • Information

9 of 22

Check In

  • Take a photo of your meal
  • Log blood pressure
  • Log medicine use
  • See how many steps walked

10 of 22

Steps Walked

  • User can earn gold based on steps walked per day
  • Steps walked is updated in real-time when walking

11 of 22

Log Medicine and Blood Pressure

  • User can enter blood pressure and medicine use on a daily basis
  • User is able to submit each information

12 of 22

Watch Videos

  • Unlock
  • In-video quiz

13 of 22

Memory Game

  • Test study outcome
  • Unlock higher level
  • Earn gold
  • Or even lose gold

14 of 22

User Information

  • User is able to edit his/her info
  • This is for research purporses only

15 of 22

Pet

  • Status
  • Play with
  • Feed
  • Clean

16 of 22

Pet Status

  • Neutral
  • Happy
  • Sad
  • Sick

17 of 22

Possible Death

  • If you do not keep yourself healthy, and away from getting high BP

18 of 22

Image detection

  • User take a picture of their meal, the app detect if it is a healthy food and give reward to user
  • Ultimate goal: An image recognition machine learning model that classifies healthy food and unhealthy food (Hard)
    • Implementation for now: AWS Rekognition + WordNet NLP
    • Rekognition detects all labels of the objects in pictures (plates, cups, napkins)
    • NLP method(WordNet) to filter out the non-food part and word more similar to fruit & vegetable
      • Wu & Palmer Similarity Algorithm

19 of 22

Take Picture of Meals

  • Amazon Rekognition
  • Natural Language Processing

20 of 22

Whats Next?

  • Successfully publish to Apple Store (in progress)
  • Collect more kinds of user data
  • Provide a better way for researchers to analyze the user data
  • Machine Learning Recognition Model
  • Customizable UI
  • Interactions between users?

21 of 22

Questions?

22 of 22

Thank You!