The Use of Logistic Regression Model in Estimating the Probability of Being Affected By Breast Cancer Based On the Levels of Interleukins and Cancer Marker CA15-3
Logistic Regression
K-Means
Classification
Breast Cancer
Breast cancer has got much attention in the recent years as it is a one of the complex diseases that can threaten people lives. It can be determined from the levels of secreted proteins in the blood. In this project, we developed a method of finding a threshold to classify the probability of being affected by it in a population based on the levels of the related proteins in relatively small case-control samples. We applied our method to simulated and real data. The results showed that the method we used was accurate in estimating the probability of being diseased in both simulation and real data. Moreover, we were able to calculate the sensitivity and specificity under the null hypothesis of our research question of being diseased or not.
