Inspiration
At the age of 15, I lost my beloved grandpa to heart disease, and it was a life-changing experience for me. Witnessing the devastating effects of this disease on my grandpa and my family, I was filled with a strong desire to take action and make a difference. However, being a computer science student, I wasn't sure how I could help in the fight against heart disease.
Fortunately, with the rapid advancement of artificial intelligence and machine learning, I saw an opportunity to apply my skills to tackle this issue. I decided to work on a project that would utilize these technologies to create a heart disease predictor, a tool that could help people assess their risk of heart disease and take preventive measures to avoid it.
What it does
The Heart Disease Predictor is a machine learning-based model that takes in various health parameters such as age, sex, chest pain type, blood pressure, cholesterol level, and more, to predict the likelihood of heart disease. It can be used as a personal assistant to help people take control of their heart health and make informed decisions about their lifestyle and medical care. By providing accurate and reliable predictions, the Heart Disease Predictor can potentially save lives by detecting heart disease early and encouraging preventative measures.
How I built it
The Heart Disease Predictor was built using Python programming language and various machine learning libraries such as Scikit-learn and pandas. I first gathered a dataset containing various health parameters of patients who have and do not have heart disease. and then preprocessed the data by handling missing values.
Next, I used the RandomForestClassifier algorithm to train the machine learning model on the preprocessed data.
After training, the model was saved using the joblib library. For making predictions, user inputs their health parameters which are then processed to create a data frame similar to the original dataset. The saved model is loaded and used to predict the likelihood of having heart disease based on the provided inputs.
Challenges I ran into
One of the most difficult hurdles I encountered was our initial attempt to make the bot a voice assistant. Despite my best efforts, I was unable to do it. Another challenge was finding a reliable and accurate dataset to train our machine learning model on. Medical data is very difficult to come by. After a lot of time, I discovered a high-quality dataset that met project needs.
Accomplishments that we're proud of
Creating a working model of this which has an accuracy of 81.1% is the biggest achievement for me despite all others.
What we learned
Learned about how to use pandas, load datasets, preprocess data, replace missing values, and finally about random forest algorithm and a lot of libraries that I don't even know existed.
What's Next for Heart Disease Predictor
Looking ahead, we learned that there are many ways to improve and expand the Heart Disease Predictor application. For example, we could create a mobile app that allows users to upload their health reports and receive personalized exercise recommendations based on the severity of their disease. We could also integrate automatic scheduling of appointments with healthcare professionals.
Log in or sign up for Devpost to join the conversation.