Inspiration
Our inspiration Often times, it gets frustrated to present symptoms and not be able to find the best prognosis on Google.com. Thus, our inspiration for MyDoc was to create a platform that could help people quickly and easily get a prognosis for their health issues, without having to wait for appointments or go through the hassle of scheduling visits to healthcare providers, or the frustration of reading articles on Google.
What it does
MyDoc is a web application that uses machine learning( RandomClassifier) to predict the disease based on the symptoms provided by the user. Users can enter their symptoms and personal information and the application will use a pre-trained machine learning model to predict the most likely disease, and using the google maps API, it will display hospitals nearby you that take care of similar medical conditions.
How we built it
We built MyDoc using a MERN stack, which includes React js for the frontend, Node.js and Express for the backend, and MongoDB(Parse on Back4app) for the database. We used Python for the machine learning component of the application and created an API endpoint using Flask to serve the predictions. We also used the Google Maps API to display the hospitals nearby
Challenges we ran into
Challenges: Challenge1: One of the biggest challenges we faced was integrating the Python machine learning model with the Node.js backend. We also had to deal with issues related to CORS (Cross-Origin Resource Sharing) and configuring the Flask API endpoint to properly receive and process requests.
Challenge2: Integrating the machine learning model was also challenging since our application heavily relied on JavaScript and the model was in Flask (python) .
Challenge3: Finding data that is diverse. Our idea was to add bias to our data such that the ML model gets trained on a robust data and takes into account your personal information. Unfortunately, no such data exist as of now.
Accomplishments that we're proud of
Our accomplishments: We are proud of creating a functional web application that provides value to users by using machine learning to predict the most likely disease based on their symptoms. We also learned a lot about integrating different technologies and working collaboratively as a team, thus we are proud to have learned and integrated those technologies in 24hrs!!!
What we learned
we learned a lot about the different technologies used in the project, including React.js, Node.js, Express, MongoDB, and Flask. We gained experience working collaboratively and learned how to effectively manage our time and divide tasks.
What's next for MyDoc
In the future, we hope to expand the capabilities of MyDoc to include more types of health issues and conditions, as well as to improve the accuracy of the predictions by using more advanced machine learning techniques. We plan on doing research on gathering medical data that is diverse and can include people from different ages, genders, races, and sexualities. We also hope to incorporate more features to help users better manage their health and connect with healthcare providers if needed, such as giving them info about the best hospital for the specific disease and being able to call them, or even giving users the option to edit their information. We also want to integrate a Large Language model that can analyze complex symptoms before passing them to the ML model.
Log in or sign up for Devpost to join the conversation.