Breast Cancer is a serious disease threatens many women nowadays. In this project, I develop a breast cancer diagnostic and treatment recommend system. This system will judge if a patient has breast cancer benign or malignant, if a patient with breast cancer, then the system will recommand most relevant research papers treatment options. It could be helpful for both physicians and patients.
This project contains two parts:
-
Breast Cancer Diagnostic we training Breast Cancer Diagnostic model to predict whether the cancer is benign or malignant with data from Kaggle and UCI Breast Cancer DataSet(https://archive.ics.uci.edu/ml/datasets/breast+cancer)
-
Research publications with treatment options Last Research materials system We create a dataset with research papers and stored in elasticsearch system, so we can push documents with treatment options for Breast Cancer Diagnostic system.
In Diagnostic part, I will focus on feature visualization and selection. Feature selection with correlation, univariate feature selection, recursive feature elimination, recursive feature elimination with cross validation and tree based feature selection methods are used with random forest classification. Apart from these, principle component analysis are used to observe number of components. After I train this predictive model. Once we have information about a patient, the model will predict if the breast cancer is benign or malignant. Then our recommend system will recommend most relevant reserach articles with treatment options highlighted.
#Reference
1.Breast Cancer Survivorship: A Comprehensive Review of Long-Term Medical Issues and Lifestyle Recommendations Balazs I Bodai, MD, FACS
-
VARIOUS TYPES AND MANAGEMENT OF BREAST CANCER: AN OVERVIEW Ganesh N. Sharma,1,* Rahul Dave,1 Jyotsana Sanadya,2 Piush Sharma,2 and K. K Sharma3
-
SF3B1 mutations constitute a novel therapeutic target in breast cancer Sarah L Maguire,1,† Andri Leonidou,1,2,† Patty Wai,1,2,† Caterina Marchiò,2,3 Charlotte KY Ng,3,4 Anna Sapino,2 Anne-Vincent Salomon,5,6 Jorge S Reis-Filho,3,4 Britta Weigelt,3,4 and Rachael C Natrajan1,2,*