This project aims to allow the users to self diagnose themselves.Vitaliser is an app created with the aim to help people diagnose their diseases by giving inputs about their symptoms in a simple questionnaire and receiving their health reports. This app will tell the acuteness of the risk your body is at and will guide you to what you next step should be.This app is made with the easiest to understand UI yet has all the features to best diagnose the disease and help the patient take a quick action in case of high risk.After the completion of the hackathon we further plan to extend this project to include more diseases in the diagnoses range and enhance the features for more better user experience.
There mainly two division of files in this project. They are: the iOS App Code files and the Deep Learning Model Files. The Deep Learning Models can be found under the DL Models Folder. The Rest of the app code can be browsed through the repo.
This app can only run on an iOS device. To run the application on Mac, clone this repository through the master branch into the desired folder and start this project on XCode. Build the project and then hit the run button.
There are two Deep Learning Models used in this project. One to detect chances of contacting Novel Coronavirus and another to detect chances of heart related health issues. The Covid Model uses a 4-layer ANN Architecture and the Heart Model uses a 5-Layer ANN Architecture. The inputs for both the models are simple day to day questions whose answers are readily available to a user.
The two models can be visualised as follows:
The Covid Model
The Heart Model
Any user view can their report of the conducted assessments in a very readable manner depicted below.
