Who says Girls can't Fight

Inspiration

"When we embrace uncertainty, it can be very liberating. If you can accept the uncertainty, it allows you to live life every day"

Breast cancer is the malignant tumor (a tumor with the potential to invade other tissues or spread to other parts of the body) that starts in the cells of the breast.

Cancer is a disease that affects millions across the globe. Just the word strikes terror into our hearts and souls. Even though a lot of us don’t know much about the disease, we associate cancer with death, sorrow, and pain.

However, there are many people who have fought cancer and emerged victorious. Many women across the world have.

What it does

There are two types of tumor-

• Malignant
• Benign Malignant tumors are the dangerous ones. They can turn into Cancerous tumor.

I have created a model using python and machine learning to help doctors around the world to predict and classify whether the tumor is Malignant and Benign depending on the features such as radius, lump thickness, etc.

How we built it

So to build this model I used dataset provided by https://archive.ics.uci.edu/ml/datasets/Breast+Cancer+Wisconsin+(Diagnostic) that can accurately classify a tumor as benign or malignant based on several observations/features.

The technique can rapidly evaluate breast masses and classify them in an automated fashion. Early breast cancer can dramatically save lives specially in the developing world.

Challenges we ran into

The challenges were very real, in healthcare field the accuracy has to be maximum and we can't have any scope or very less scope for errors. So to improving the accuracy of the model was the main challenge. At the end I could finally increase the models efficiency and accuracy to 97 % and the 3% of the error that had in my model was type-I error, i.e the cancer was benign and it stated it as malignant which is sort of safe as compared to type-II(cancer being malignant and shown benign, that's more risky)

Accomplishments that we're proud of

The experience To develop my very first healthcare model and to complete it with so much efficiency was really an amazing thing for me.

What we learned

While doing the research on this project I got to know a lot about what women go through in this, I learned a lot while testing and training datasets such as feature scaling, improving the model and much more.

What's next for Who says Girls can't Fight

The technique can be further improved by combining computer vision/ML techniques to directly classify cancer using tissue images.

Built With

• machine-learning
• python
• sci-kit