摘要
大数据对肿瘤临床研究有非常重要的意义,骨肿瘤病理亚型众多,且发病率均不高,因此导致单一医疗机构的病例数据量比较少,目前国内一些大型的骨肿瘤诊治中心也建立了一定规模的数据库,如积水潭医院的骨肿瘤数据库[1],然而全国范围的骨肿瘤登记系统及大数据库仍然是空缺。本期刊出的柳昌全等[2]回顾性分析 1995 年至 2014 年,来自 SEER 数据库 1910 例四肢骨肉瘤患者,通过单因素 Log-rank 分析和多因素 COX 分析。
Low incidence of bone tumors results in the lack of high-level evidence-based clinical data.Retrospective studies based on multi-center,multi-regional,even national database can provide valuable information to guide future clinical research.Scholars use the open public database SEER to extract the information of patients with osteosarcoma of extremities and to establish a new prognosis prediction model,which will effectively integrate independent prognostic factors and accurately predict the prognosis.This model is an attempt to use big data diseases for analysis and provide more reliable conclusions than the research of small sample size,providing help for clinical decision-making.Some prediction models will be established on the basis of big data and in-depth learning,and further provide references for clinical practice with broad application prospect.
作者
华莹奇
HUA Ying-qi(Department of Orthopedics,Shanghai General Hospital,Shanghai,200443,China)
出处
《中国骨与关节杂志》
CAS
2020年第8期561-562,共2页
Chinese Journal of Bone and Joint
关键词
骨肿瘤
大数据
临床研究
预测
模型
生物学
Bone neoplasms
Big data
Clinical study
Forecasting
Models,biological
