Inspiration
As a teenager looking back at the generation before us there is so much rush in our lives we often forget to keep track of any healthy habits and even if we try we fail.
What it does
Lifestyle Tracker :
- Sleep tracking and logging with dates to analyze sleep schedule.
- Stress tracking with multiple sentiments to not ignore mental health.
- Graph for better understanding of analysis.
- Breathing exercise.
- Community to share thoughts and what worked for you.
Disease Diagnosis :
- Multiple symptoms with easy input.
- Huge dataset to diagnose disease from.
- NLP to understand symptom inputs better.
How we built it
Lifestyle Tracker: We built this webapp using MERN stack. And used python and python libraries to implement logic behind sentiment analysis. Disease Diagnosis : Developed using python, python libraries, jupyter notebook. Used multiple and complex machine learning algorithm to predict diseases from known symptoms. Used dataset from kaggle for better results.
Challenges we ran into
Most of the challenges we faced were with Disease Diagnosis app, where we had to train a very huge dataset and had to have the most accurate model possible since health is a sensitive subject. But due to this complexity our model exceeded resource usage. Therefore it became impossible for us to host our app (at free hosting sites).
Accomplishments that we're proud of
Developing such a big project in such a short time span and trying to solve an actual problem is something of an accomplishment for us we consider.
What we learned
Practicing from tutorials and actually working on a project from scratch is really a impactful experience to have/
What's next for Swasthya
Will try to host this website and make it available to users as soon as possible. Collect users data securely and encrypted, to automate these small tracking of the day making it non noticeable but reliable for results.
Built With
- express.js
- jupyter
- keras
- mongodb
- natural-language-processing
- nltk
- node.js
- pandas
- python
- react
- tensorflow
Log in or sign up for Devpost to join the conversation.