Cataract Detection

Helps to determine whether a person has cataract or not based on symptoms.

Comment

Inspiration

Cataracts are the biggest cause of vision loss in the world and traditional statistical methods are incompetent in discovering hidden links from numerous medical records, whereas data mining and machine learning are the most promising techniques, that can tackle this problem. The concept of data mining along with efficient machine algorithms is to be used in the creation of our model, with further models to be added as per requirement. With the world becoming more connected than ever before, there sure isn't going to be a shortage of data production. Medical researchers can use large amounts of data on treatment plans and recovery rates of cataract patients in order to find trends and treatments that have the highest rates of success in the real world.

What it does

The user simply has to fill a basic details form, which mainly comprises of various symptoms of cataract. Based on the information provided we can run various algorithms and predict whether that person has chances of developing cataract or not.

How we built it

For Front End we used HTML5, CSS3, Bootstrap, and Javascript. For Back End we used google colab for working on our model.

Challenges we ran into

We were able to train and test our dataset using three different algorithms(Naive Bayes, Decision tree, Multi Layer Perceptron) but we were unable to connect our front end to the back end.

Accomplishments that we're proud of

We successfully trained and tested our model with an accuracy of 89%

What we learned

We learned about various libraries in python, various classification algorithms, and how to design a web-page using html, css, bootstrap and java script

What's next for Cataract Detection

We can create a application which the user can download directly on the phones, thereby making it more convenient for the user. We can also implement an image processing model which can detect cataract using the images of the eye of patients.

Share this project:

Updates