Breast Cancer Prediction
Machine Learning Project
Meet Our Team
Fatma Butun
Jenny Yang
Matt Keeley
Shay Masood
Shrey Saraiya
Project Overview
Breast cancer is one of the most common cancers among women globally, resulting about 12% of women affected and the number is still increasing, making it a significant public health problem in today’s society.
Breast cancer arises when cells in the breast start to develop abnormally. If not identified in the early-stage, it could be fatal to the patient. It is the most common type of all cancers and also the main cause of women’s death worldwide.
Nowadays technologies and data mining methods are making the prediction and classification work more efficient and effective. Machine learning is widely used in medical field, helping with diagnosing various diseases.
This project utilizes Machine learning to train various models and select the one that can best classify which type of cancer the patient might have. A prediction portal is built to take measurement of key features and return an initial classification which could potentially help doctors to provide timely treatment to patients and improve the chance of survival.
Data Source:
https://www.kaggle.com/uciml/breast-cancer-wisconsin-dataBreast Cancer Facts
The Breast is a gland consisting of fibrous and fatty tissues of various amounts resting on the pectoral muscle and surrounded by the skin.
The breast has an extensive morphological and functional variability. It is divided into several parts with different characteristics: the skin, the nipple and areola, the fatty or adipose tissue, and the breast tissue proper or corpus mammae. The breast tissue is organized in a strict lobular scheme, with almost no interconnection between lobes.
Each lobule is lined with lactiferous duct cells surrounding the center of the duct that opens on the nipple.
The development and homeostasis of the breast morphology is based upon initialization and branching of the buds and ducts, and development of the lobules. Lobulization is the step that is most sensitive to hormonal change, and is easily altered.
Prediction
Link to the prediction portal:
https://bc-predict.herokuapp.com