Hotei - Wellness Activity Recommendation

1. Overview of Hotei

The issues of stress and general emotional well being are important in the context of student life where almost 63% suffer from stress, and 27% of all students suffer from a mental problem of some sort. This project aims to help manage a student’s stress levels whilst helping them become more aware of their emotional well being. This will be done through the development of a smartphone app, making use of heart rate data from a wearable device to automatically detect moments of stress, a recommendation system to provide tailored activity suggestions to help them relieve such stress, and ultimately improve their over all emotional well being.

2. Hardware

• iPhone (iOs 10+)
• Polar H7 Heart Rate Sensor
• Apple Watch (Optional)

3. Files

Files under /Hotei (not all are shown):

The main project source code resides here.

• Hotei.xcworkspace: Entry point for iOS project. (Open this instead of .xcodeproj)

• Hotei/: Subdirectory holding iOS app development files

  • Main.storyboard: UI Story board of iPhone

• HoteiW/: Subdirectory holding WatchOS app development storyboard

  • Interface.storyboard: UI Story board of Apple Watch

• HoteiW Extension/: Subdirectory holding WatchOS app development files

  • InterfaceController.swift: Watch logic

• Pods/: Subdirectory holding CocoaPods

  • OpenCV/: Subdirectory holding OpenCV lib
  • Charts/: Subdirectory holding Charts lib

• Podfile: CocoaPod setup file (Instrustion for CocoaPods to install external lib correctly)

Files under /Database (not all are shown):

The database source code resides here.

• HoteiApi/: Subdirectory holding database source code
• HoteiClientExample/: Subdirectory holding database example code
  • HoteiClientExample.csproj/: Database API call example

Files in the root directory (not all are shown):

Auxiliary files go in here.

• Coursework Instructions.pdf: Goal and instruction for this project

• Data/: Subdirectory holding training and learnt data:

  • survey_results.ods: Training data for recommendations
  • user?.csv: Users heart rate data
  • user?_test.csv: Test data for users

• figures/: Subdirectory holding figures used in the report:

  • System-overview.jpg: High-level view of the ststem
  • sysdia.jpg: Flowchart of the Feedback State
  • database.jpg: Database relationships
  • context.png: Context fitering
  • conf_user.PNG: Confusion matrix
  • screen.png: Screenshots of the app on iPhone and AppleWatch

• Reports/: Subdirectory holding reports:

  • DesignReport.pdf: First stage design report
  • Individual Component Report - *.pdf: Individual reports
  • FinalReport.pdf: Final report

• README.md: This file (readme)

4. Dependency

Notice that the project uses CocoaPods for installing external library. Please have it installed in advance.

• Charts: Charts lib for displaying charts
• OpenCV: OpenCV lib for machine learning

5. Running Hotei

Installing CocoaPods:

$ sudo gem install cocoapods

Clone this project

$ git clone https://github.com/timkchan/Hotei.git

Install dependencies (in repo root dir):

$ cd Hotei
$ pod install

Open Xcode project (in repo root dir):

$ cd Hotei
$ open Hotei.xcworkspace

Project uses CoreData and external libraries. It's recommended to clean and rebuild the project. Now the app should be ready to run on a simulator or an actual device.

6. Demo video

• Youtube: View Video
• Download: Download Video

7. Contribution

This is a 4th year class (EE4.67 – Mobile Healthcare and Machine Learning) project at Imperial College London supervised by prof. Yiannis Demiris

Team members:

• Tim K. Chan
• Ryan Dowse
• Akshay Garigiparthy
• Sagar Patel
• Nick Robertson