To access the app, click the following link: Breast Cancer Prediction Web App. You can use this web app to input a patient's details and receive a prediction on whether the patient has breast cancer.
Details for Milestone are available on Canvas.
Our project aims to build a predictive model that can accurately diagnose a breast tumour as either malignant or benign based on the provided features. The importance of this type of model lies in the early detection of breast cancer, which can significantly improve the chances of successful treatment via early intervention. We are interested in this dataset because breast cancer is a globally prevalent and devastating disease, and any efforts to improve the accuracy of its early detection are significant. In addition, we wish to explore the relationship between different features of cancer and its diagnosis to determine the most important predictors. We are curious to see how well the model performs compared to human diagnosis accuracy and whether it would be feasible to use such a model in a clinical setting. It is possible to build a user dashboard with this dataset that would allow medical professionals to input patient information and receive a prediction of the likelihood of the breast tumour being malignant or benign.
The Breast Cancer Wisconsin (Diagnostic) Data Set consists of measurements of breast cancer cases that Dr. William H. Wolberg, a medical professional at the University of Wisconsin Hospitals, Madison, gathered to study trends between benign and malignant cancers. Dr. Wolberg was accompanied in this endeavour by W. Nick Street and Olvi L. Mangasarian from the Computer Sciences department of the university. The data was initially collected in 1992, with the diagnostic version of the dataset being compiled in 1995. To record this data, a Fine Needle Aspirate (FNA) of a breast mass was taken, and features were calculated from that image. The data set has a diagnosis label of either benign (B) or malignant (M) for 569 cases in 30 data columns produced from digital images of an FNA of a breast lump. The hospital treated the patients who underwent FNAs, and the photographs of the vast majority were digitized. To assess the mass's severity, the masses' features were computed. The general public will benefit from understanding what cancer forms might resemble through this data set.
- Mithish Ravisankar Geetha: I am a third year CS major passionate in aviation, artificial intelligence and mathematics.
- Rajveer Sodhi: I am a third year COSC major, DATA minor. I love programming, movies, and astronomy!
- Yash Kumar: A Data Science major passionate about machine learning, emerging technology, and cricket.
Link to the Original Dataset: https://archive.ics.uci.edu/ml/datasets/Breast+Cancer+Wisconsin+%28Diagnostic%29