Predicting a second breast cancer using patient level data

Background: Breast cancer had an age-standardised incidence rate of 110.9 per 100,000 person-years in the female population in Switzerland in the period 2011-2015. About 7.43% of women who had been diagnosed with an initial primary breast cancer will have a second primary breast cancer within 10 years. Being able to predict a future second breast cancer for women with a first breast cancer using patient level data is a valuable tool for doctors. Early detection, already from the first breast cancer incidence can allow them to possibly prevent a second breast cancer event, detect it earlier or choose the most beneficial available treatment. Machine learning as well as deep learning algorithms are vastly applied to the field of medical health and sometimes they work as a support system to case-based reasoning to help doctors improve accuracy of diagnostic and prognostic decisions or to enhance the performance of Computer Aided Diagnosis (CAD) systems. Aim: The aim of this project is to build a prediction model for second breast cancer for women who have had a primary breast cancer in the past. The model should be able to classify a women with a first breast cancer as a high or low risk for a second breast cancer.