Accelerated Failure Time Models with Applications to Endometrial Cancer Survival Data

The objective of this study is to determine the significant predictors endometrial cancer using accelerated failure time models (AFTM). We have demonstrated applications AFTM viz. Exponential, Weibull, Log-normal, Log-logistic, Gompertz, Gamma and Generalized AFTM, as an alternative Cox proportional hazard model. Data for analysis was collected from Acharya Harihar Post Graduate Institute Cancer (AHPGIC), Cuttack, Odisha during period 2016–20. Based on lowest Akaike Information Criterion (AIC) Bayesian (BIC) value, Weibull has been chosen best fitted AFT such age, comorbidity, tumor size, isolated para-aortic adnexa found (p-value < 0.05) explain survival patients. Hence, by optimizing different treatments, based prognostic factors plays important role in managing at early stage.