Brain Stroke Prediction Using SVM

Facial drooping, Arm weakness, Speech difficulties and Time to call emergency services

Comment

Inspiration

As per WHO nowadays getting STROKE is the 2nd leading cause of death globally and also it's responsible for approximately 11% of the total deaths globally. Due to these reasons, we felt it was appropriate to develop a solution for this problem and help people predict the symptoms or cause of stroke before itself. We also referred to some projects in this category and decided to do that, but way better.

What it does

As of now, we are just developing a model to predict the stroke of the patient. The Defined model predicts whether a patient is likely to get a stroke based on the input parameters like gender, age, various diseases, and smoking status of the person.

How we built it

We have built an ML model using SVM as well as KNN where we have compared both models but as per our observation we found that the SVM model is giving the most accuracy so by default we can use SVM but for reference, we are showing both models in our Google-Collab notebook link. We also showed the visualization of the data using different types of graphs by using seaborn, plotly, and pyplot.

Link for Dataset used

DATASET_LINK

Challenges we ran into

As we are a beginner in AI/ML we have used most of the technologies for the project to give our best. So, as they are new to us it took time for us to understand and implement. As our team collaborated and worked it became easy for us to implement the project perfectly.

Accomplishments that we're proud of

Now, using the inputs required the people can easily predict if there is a chance for them to get a STROKE or not.

What we learned

We have learned various technologies, the use of platforms, comparing between models, knowledge of AI/ML, experience in developing projects, and so on.

What's next for Brain Stroke Prediction Using SVM & KNN

Further, we can implement the user input the input parameters so that we can take input from the user and predict the cause of stroke. We will also try to remove outliers and then retrain the model to check for a decrease in misclassification.

Built With

  • collab
  • knearest-neighbors
  • matplotlib
  • python
  • seaborn
  • support-vector-machine
Share this project:

Updates