Prediction the Onset of Diabetes

Image of output obtained
Website image

Inspiration

Being Ai students and working on Machine learning algorithms, we ventured into the health industry and came across blogs and resources on diabetes explaining the dangers it poses. All of us have heard of the term 'Diabetes' since childhood and it has become a common term nowadays but the threat it poses hasn't decreased a bit. The trends in current lifestyle have led to the most significant increase in cases of diabetes and we decided it would be a good idea to make an ML algorithm that could predict the onset of diabetes in an individual by analysing their health and helping them make better lifestyle choices to avoid this dreaded common disease.

What it does

The website that we made will help people to predict the onset of Diabetes in the future by taking some inputs based on concurrent body conditions. Some examples are - BMI, Insulin resistance etc. Its been trained using the Pima diabetes dataset which was collected from more than 700 people in the U.S.It analyses the dataset and trains the model based on it. Which then is used to predict the onset of diabetes.

How we built it

The building process of the model started with the collection of the dataset, Cleaning it, Removing the null values in the dataset, and then using a test train split function split the dataset into testing and training datasets which in turn were used to test and train our model. The training result that we got used to predicting real-life conditions. Using Gradio for the interface made the work so easy and convenient as the integration of the model to the interface was fairly simple work to do.

Challenges we ran into

we had to create a interface/website for this project so had to try different module and apps for creating it finally we were able to create it by using Gradio

Accomplishments that we're proud of

We could get decent accuracy in our ML algorithm as well as we were able to create an interface/website to make our project more user friendly

What we learned

We were able to learn more about Machine learning and also on how to create a interface/website for same as well. We also learned somethings on the topic of diabetes

What's next for Prediction the Onset of Diabetes

With large and more accurate dataset we shall be able to get a better accuracy and also a better classification algorithm can make give a better accuracy and result ,which would further improve the prediction as well as the detection of the onset of diabetes