基于307例宫颈癌患者临床特征的生存状态预测分析  

Prediction Analysis of Survival Status Based on Clinical Characteristics of 307 Cervical Cancer Patients

作　　者：孙鹏哲[1] 许春洁 闫祖威[1] 冯永娥[1] SUN Pengzhe;XüChunjie;YAN Zuwei;FENG Yonge(College of Science,Inner Mongolia Agricultural University,Hohhot 010018,China)

机构地区：[1]内蒙古农业大学理学院,呼和浩特010018

出　　处：《内蒙古大学学报(自然科学版)》2025年第1期69-76,共8页Journal of Inner Mongolia University:Natural Science Edition

基　　金：国家自然科学基金项目(62262050,32160258);内蒙古自治区教育科学研究“十四五”规划课题项目(NGJGH2024053)。

摘　　要：通过探索宫颈癌患者的临床特征与生存状态之间的关系,构建准确可靠的生存状态预测模型,可以为宫颈癌的预防、诊断和治疗提供新的思路和方法。本文基于TCGA数据库307例宫颈癌患者的临床随访数据,运用Spearman相关性分析探讨宫颈癌临床特征的相关性,首次运用因子分析探究了宫颈癌临床特征的基本结构,并提出基于因子分析的二元Logistic回归生存状态预测模型。模型预测总精度为84.2%,召回率为84.2%,精确度为81.0%,模型的AUC值为0.861,该模型在分类任务上表现良好,具有较高的预测能力和可靠性。By exploring the relationship between clinical characteristics and survival status of cervical cancer patients,and then constructing an accurate and reliable survival status prediction model,new ideas and methods can be provided for the prevention,diagnosis,and treatment of cervical cancer.Based on the clinical follow-up data of 307 cervical cancer patients from the TCGA database,Spearman correlation analysis is used to explore the correlation between the clinical features of cervical cancer,and innovatively employed factor analysis to investigate the basic structure of these clinical features.A binary logistic regression survival state prediction model based on factor analysis was proposed.The model's overall prediction accuracy was 84.2%,the recall rate was 84.2%,the precision was 81.0%,and the AUC value of the model was 0.861,indicating that the model performed well in classification tasks with high predictive capability and reliability.

关 键 词：宫颈癌 生存状态 因子分析 LOGISTIC回归 

分 类 号：Q61[生物学—生物物理学]