Inspiration

In this bleak and confusing corona and home ridden times, the entire fear is caused due to limited information and the rule of misinformation is prevalent. The lack of coordination and proper information has caused so many deaths. So we thought, we will develop a symptom analyzer that can accurately predict what kind of disease is the patient suffering from. This can help doctors speed up analysis and also help in predicting an outbreak! It can also inform the public about what kind of disease they might be facing which helps them to understand if they need to go to a doctor immediately to get a proper diagnosis. This can help the person to get the required treatment on time.

What it does

Prognosist uses machine algorithms to identify the possibility of a disease by analyzing the symptoms. This can help doctors improve the speed and accuracy of their diagnosis. Doctors can also use our system to find out rarely occurring diseases that might be difficult to diagnose. Currently, we can analyze 132 symptoms and predict 47 diseases. This can be improved by adding more data to the dataset. We have partially added an option that allows doctors to add their diagnosis to our dataset and help us improve it.

How we built it

The entire web application is built using Django. We used Jupyter notebook to run analysis and trained our model using the Random Forest Algorithm. Dataset was taken from Kaggle. link . Frontend is built using simple HTML, CSS, and javascript. A minimal user-friendly interface is used to make operations easy. The model is also saved directly using pickle and is not trained every time to increase the efficiency

Challenges we ran into

Getting a good dataset was one of the main issues we faced while developing this project. There were very few datasets with sufficient information. This created a huge problem for us. We managed to find a good one on Kaggle after searching lot. The dataset was not complicated and allowed us to train a model easily. We also faced certain development issues with Django which we managed to overcome.

Accomplishments that we're proud of

We were able to predict diseases with high accuracy. We believe that our project has huge potential in improving medical field. It can save many lives by giving doctors the correct diagnosis at the fastest rate possible.

What we learned

Building web applications is easier on Django. We learned how to use Django and also covered concepts of machine learning that can help us in our career.

What's next for Prognosist

The application can further be improved to predict an outbreak. Our database provides sufficient information for such a prediction. By analyzing the rise of symptoms in a particular location, we can detect if there is an outbreak. This can help us stop another pandemic. This application has a long way to go and can revolutionize the medical industry

Share this project:

Updates