Overview of My Submission

Cardiovascular diseases normally represents a condition that involve narrowed or blocked blood vessels that lead to the heart failure, attack or stroke. It is one of the biggest mortality among the population of the world.
Cardiovascular diseases (CVDs) are the number 1 cause of death globally, taking an estimated 17.9 million lives each year, which accounts for 31% of all deaths worldwide. Four out of 5CVD deaths are due to heart attacks and strokes, and one-third of these deaths occur prematurely in people under 70 years of age.

So, people who have high risk of heart disease/ heart failure or also those who want to check, requires early detection and proper management of their disease. For this purpose, I developed the Heart Care App in which user will be classified whether he has heart disease or not.

Submission Category:

My own adventure

Link to Code

My Github Repository for Flutter App
My Github for fastAPI backend
Github for Colab Repository

Images

Images are loosing resolution so these can also check in the Readme.md of HeartCare app

Calculating beat video

Video which shows how heart beat is calculating

What it does

Currently, user have to make an account. A dashboard is availabe on which user monitoring of heart beat and daily heart disease prediction is showing. Downwards, medical history cards are present if user want to check his history. Second screen is to add medical record in which user has to fill few details like blood pressure, sugar, age, gender. One more button is present which is takes the user to calculate heart beat screen. On heart beat screen, user have to place finger on mobile camera and torch. Then using the photoplethysmogram technique, user's heart beat is recorded. Then user can save medical record. On successfully saving the record, user will be notified that whether he has heart disease or not. A firebase cloud messaging is also integrated which daily on 8PM generates a push notification that check your heart.

How i built it

MongoDB - Obviously the best NoSQL database emerging rapidly.
Flutter
Fast API
Machine Learning
Flutter charts

Challenges

I faced a lot of challenges in creating this app. First, I am not familiar with state management in flutter. So I got lot of errors in integrating it. Secondly, I have never used MongoDB Realm. So, I tried to generate push notifications using mongoDB Realm but it didn't work because few libraries are not supported when i tried to install node-fcm. Also MongoDB Realm SDK is not supporting Flutter yet. It is in progress. Then, I am not familiar with Data analysis and machine learning. So, lot of errors, bugs and functionalities are my challenges.

Accomplishments I'm proud of:

I am proud that all the above technologies which are new for me, have learnt. The most important purpose of participating in MongoDB Atlas Hackathon is to learn all the technologies. I am proud that i have learnt FastAPI, MongoDB Realm, Flutter Provider State Management. This hackathon becomes most useful for me.

What's next:

• More functionalities
• Validation of fields
• More Charts
• Also Make for doctors and patients (separate interface)

Team

1- Hina Khadim
My LinkedIn Connect

Thanks to all, especially MongoDB Team for organizing such a great Hackathon in which we all have learned new technologies, frameworks, explore amazing features of existing technologies. Expecting more hackathons in Future.