Breast-cancer-prediction

Context This dataset includes data from a random sample of 20,000 digital and 20,000 film-screen mammograms received by women age 60-89 years within the Breast Cancer Surveillance Consortium (BCSC) between January 2005 and December 2008. Some women contribute multiple examinations to the dataset. Data is useful in teaching about data analysis, epidemiological study designs, or statistical methods for binary outcomes or correlated data. Content The data set contains 39998 rows and 13 cols. Attributes are described as follows: Field Name Type (Format) Description AgeAtTheTimeOf_Mammography number Patient's age in years at time of mammogram Radiologists_Assessment string Radiologist's assessment based on the BI-RADS scale --- --- --- IsBinaryIndicatorOfCancer_Diagnosis boolean Binary indicator of cancer diagnosis within one year of screening mammogram (false= No cancer diagnosis, true= Cancer diagnosis) --- --- --- ComparisonMammogramFrom_Mammography string Comparison mammogram from prior mammography examination available --- --- --- PatientsBIRADSBreastDensity string Patient's BI-RADS breast density as recorded at time of mammogram --- --- --- FamilyHistoryOfBreastCancer string Family history of breast cancer in a first degree relative --- --- --- CurrentUseOfHormoneTherapy string Current use of hormone therapy at time of mammogram --- --- --- Binary_Indicator string Binary indicator of whether the woman had ever received a prior mammogram --- --- --- HistoryOfBreast_Biopsy string Prior history of breast biopsy --- --- --- IsFilmOrDigitalMammogram boolean Film or digital mammogram (true=Digital mammogram, false=Film mammogram) --- --- --- Cancer_Type string Type of cancer --- --- --- Acknowledgements We acknowledge the Breast Cancer Surveillance Consortium (BCSC) for making this data set available for research purposes.