Survival Analysis of Breast Cancer Patients: A Population-Based Study from SEER

Breast cancer is among the most widespread diseases in women, and it has a high rate of fatality. Accurate estimation of the future risk in patients is of vital importance. It can guide clinicians in choosing the best course of treatment. For the clinical decision-making process, it is crucial to comprehend how various factors affect patient survival. In this regard, we conduct survival analysis on a group of breast cancer patients. The data is collected from the SEER study conducted by the National Cancer Institute (NCI). For risk prediction in cancer patients, we used the Random Survival Forest (RSF) method. The Kaplan-Meier estimation was utilized to analyze the effect of different feature variables on the survival of patients. The hazard associated with the features is calculated using the Cox model. The most significant predictor variables behind the progression of cancer have also been identified. The proposed RSF method produced a c-index value of 0.752 outperforming other survival analysis models like the Cox regression model or the Gradient Boosting method. The comprehensive analysis conducted in this study can be beneficial in breast cancer research and provide insights to guide clinicians in decision-making.