This project is done as a part of the course - Introduction to Deep Learning offered in SPRING 2018. Code was made in Python from scratch as well as using Tensorflow as framework. We have build numerous architectures varying hyperparameters and also gave our intuitions onto why we got those results.
Preview: Breast cancer is the most common female cancer worldwide representing nearly a quarter of all cancers with an estimated 1.67 million new cancer cases diagnosed in 2012. Breast cancer is sometimes found after symptoms appear, but many women with breast cancer have no symptoms.Early detection means finding and diagnosing a disease earlier than if you would waited for symptoms to start.
DataSet: http://archive.ics.uci.edu/ml/datasets/breast+cancer+wisconsin+(diagnostic) This is the link for the dataset we made use of in this project. It's called the Breast Cancer Wisconsin (Diagnostic) Data Set. It contains info of various features ranging from perimeter of the tumor formed to its compactness.
Goal : To build a Convolutional Neural Network (CNN) to classify the diagnosis as Malignant or Benign outperforming human accuracy.