论文部分内容阅读
这项研究的目的是采用参数模型确定影响结直肠癌患者生存时间的影响因素,并且通过预测误差选择最好的模型。预测的准确性通常是通过损失函数测量的,也被称作预测误差。Baghestani等选择了600例结直肠癌患者,跟踪随访至少5年,为这项研究提供了完整的信息。分析因素包括体质量指数(BMI)、结直肠癌家族史、性别、肿瘤位置、肿瘤分期等因素。生存情况的单因素分析采用log-rank检验,多因素分析采用参数模型,包括log normal、Weibull和log logistic回归模型,通过预测
The purpose of this study was to use parametric models to determine the influencing factors affecting the survival time of patients with colorectal cancer and to select the best model by predictive error. The accuracy of the prediction is usually measured by a loss function, also known as prediction error. Baghestani et al. selected 600 patients with colorectal cancer and followed up for at least 5 years, providing complete information for this study. Analytical factors included body mass index (BMI), family history of colorectal cancer, gender, tumor location, tumor stage, and other factors. Univariate analysis of survival using log-rank test, multivariate analysis using parametric models, including log normal, Weibull, and log logistic regression models, through prediction