Final Masters Project to design and implement an application to detect stress, depression, loneliness and suicide risk, giving custom advices to users.
This project is a modern Android app that allows the detection of mental disorders such as stress, depression, loneliness and suicide risk, through the extraction of data from wearable devices and the self-assessment questionnaires.
Emotional Wellbeing provides indicators and charts of user's mental indicators, anonymized statistics on how the community is doing and, depending on the user's state, guidelines to improve user's emotional well-being and redirects, if necessary, to professional support.
- User measures visualization: As home screen, app offers briefly user status, showing today's level and advice extract.
Moreover, user can access more detailed information, like a bar chart of current week status, yesterday status and seven-days average; or the detailed advice depending on measure level.
- Biometric Data Extraction: The app integrates with wearable devices (only on Health Connect compatible devices, like Fitbit and Samsung wearables) to collect biometric data.
- Self-Assessment Questionnaires: Users can complete validated psychological questionnaires to evaluate their emotional state. These questionnaires are prompted twice a day via notifications, and they can be resumed anytime.
- Advanced stats: User can access historical data visualization in a line chart. Also, some filtering can be done: by measure, time interval and the granularity to group data.
Meanwhile, community status can be checked: for each measure, user can check yesterday, last seven days and current week averages.
-
User-Friendly Interface: The app features an intuitive design, making it accessible to users without technical expertise. Also, app is available (and configurable) on light/dark mode and with wallpaper color schema (only in devices running Android 12). Some examples:
-
Onboarding:
-
Permission management:
-
Splash screen everytime that app is opened.
-
-
Multilingual Support: App can be used in either English and Spanish. By default is used device language, but user can override this via settings panel.
- Core: Kotlin, Dagger Hilt
- Frontend: Jetpack Compose, Material Design 3, Vico, Navigation Compose and Lottie
- Backend: Health Connect, DataStore, Lingver, Retrofit, Room, SQL Cipher and Worker
- Install Health Connect app in target if it runs Android 13 or lower and you don't have it.
- For community graphs, run and configure an instance of API Service.
- Code can be compiled in Android Studio with a JDK 17. If you don't have or aren't sure about that, go to File | Settings | Build, Execution, Deployment | Build Tools | Gradle and download a JDK selecting Version 17.
Any contributions you make are greatly appreciated, so if you have any idea of how to make this project better, please create a pull request. Also if you find any bug, please create an issue.
This project was originally by @VicDominguez with the support of ETSISI-UPM (School of Computer Systems Engineering, Polytechnic University of Madrid) and @maria-marco
