An AI Based Heart Disease Predictor which uses basic questions for diagnosis.
In a populus country like India, there is always a surplus of patients and a lack of doctors. With such a scenario, it can be hard for easy and early diagnosis of killer diseases, such as cardiovascular diseases.
Our project is an AI-powered heart disease prediction system which analyses patient data and gives a diagnosis as to what is the likelihood have having said disease.
The user can visit the website to access the quick and easy way for the diagnosis of a cardiovascular disease.
They are greeted with a login page, and then redirected to the main form where they fill in simple details such as age, weight, height, heart rate, etc.
With said parameters, the model on the backend gives a percentage of the chance that the user might be having a heart disease, and whether they should consult a doctor or not.
The AI model is made and trained using the scikit-learn library of python. The model used is the MLPClassifier (Multilayer Perception, more commonly known as a neural network). We extract the data from the training file and fit it in a datatype where its able to be read by the neural network, and the model is trained and stored in a .pkl file using the joblib library in python.
The website is made using Python along with FastAPI, and for front end we use HTML and CSS.