Discipline: Data Science
Subcategory: Cancer Research
Session: 4
Room: Senate
Solangé Tchounwou - Tulane University
Co-Author(s): *, Solange S. Tchounwou 1, Clement G. Yedjou 2 , Jameka Grigsby 3 , Kearra Johnson 4 , and Paul B.Tchounwou 4
Breast cancer is the most common malignant in women worldwide. Women in the United States has a 1 in 8 chance of developing invasive forms of breast cancer during theirlifetime. Breast cancer arises in the lining cells (epithelium) of the ducts or lobules in theglandular tissue of the breast. The goal of the present study was to use ML application toundertake an extensive comparison of the invasive forms of breast cancer including, infiltratingductal carcinoma, infiltrating lobular carcinoma, and mucinous carcinoma. To achieve this goal,we used machine learning algorithms and collected scientific datasets of 334 breast cancerpatients available at https://www.kaggle.com/amandam1/breastcancerdataset and interpreted thisdataset based on the form of breast cancer, age, sex, tumor stages, surgery type, and survival rate.Among the 334 patients, 70% were diagnosed with infiltrating ductal carcinoma, 27% withinfiltrating lobular carcinoma, and 3% with mucinous carcinoma. Overall, 64 out of 334 breastcancer patients were in stage I, 189 patients in stage II, and 81 patients in stage III. Sixty-six (66)patients underwent lumpectomy, 67 patients underwent simple mastectomy, 96 patientsundergone modified radical mastectomy, and 105 patients underwent other types of surgery. Thesurvival rate was 83.4% for stage I, 79.1% for stage II, and 77% for stage III, respectively.Finding obtained from the present study demonstrated that machine learning provides large datathat holds great promise for improving breast cancer outcomes.
Funder Acknowledgement(s): This work was funded by the National Institutes of Health (NCI), Grant #1U54MD015929-01 at Jackson State University, Jackson, MS, United States, and FacultyResearch Award Program (FRAP) at Florida Agricultural and Mechanical University,Tallahassee, FL, United States.
Faculty Advisor: Dr. Clement G. Yedjou, clement.yedjou@famu.edu
Role: I helped collect scientific datasets of 334 breast cancer patients that were available at https://www.kaggle.com/amandam1/breastcancerdataset. Using this data I helped compare and interpret datasets based on the form of breast cancer, age, sex, tumor stages, surgery type, and survival rate.