Statistical Evaluation of Survival Rates in Lung Cancer Utilizing Gaussian and Logistic Regression Techniques

Cancer is the second most prevalent cause of mortality globally as per the World Health Organization. Among the various types of cancer, lung cancer is particularly fatal and ranks third in terms of frequency. Its impact on healthcare systems and individuals’ quality of life is enormous. This sort of cancer is particularly problematic because it has a very poor survival rate compared to other types of cancer. The focus of the present research is to examine the correlation between lung cancer and survival time, in addition to the different characteristics of the cancer dataset. The purpose of the present investigation is to determine the optimal modeling strategy for accurately assessing the survival probabilities and other statistical measures. A set of Gaussian and Logistic parametric regression survival models to calculate probability values, average survival time, and other relevant statistical metrics have been used in the present research study. The data of 168 patients and nine essential variables related to advanced lung cancer, including age, gender, and other clinical factors have been included in the study. The proposed estimation methods are compared by assessing significant factors, such as mean survival probability, mean cumulative survival probability, and model fit indices viz, the Akaike Information Criterion and Bayesian Information Criterion. The family of Logistic Regression models exhibited higher performance across these parameters, reflecting their resilience and appropriateness for this particular set of survival data.