The incidence, prevalence and case fatality rate of T2DM are still high, its risk factors are increasing and popular , although lots of funds have been used for the treatment of T2DM in the developed or developing countries for more than ten years. To discover early and prevent T2DM has become the global urgently needs to solve an important public health problem. Now overseas researchers employed the tool of risk evaluation to predict the onset risk of T2DM in order to find and interfere with the high-risk group to control the risk factors, thereby the occurrence of T2DM will be decreased. This study will screen the variables of model using the data mining such as Logistic regression, cluster analysis and principal component analysis and consulting the expert opinion of DELPHI, based on the prophase nest case control study of T2DM; the model of individual onset risk of T2DM will be founded by the Support Vector Machine technology used MATLAB and MATLAB SVMTOOL as the platform and integrated the influence of the gene, environmental factors and their interaction for T2DM; the sensitivity, specificity and predictive precision of the model will be evaluated by ROC; the external validity of model will be validated prospectively through the other cohort. This study intends to expand the new thread and to supply the new measure for the precaution, prevention and individual alternative health service of T2DM to reduce the burden of T2DM finally and to promote the population health.
近十多年,无论是发达国家还是发展中国家,均投入大量经费用于T2DM的治疗,但T2DM的发病率、患病率和病死率仍然居高不下,危险因素仍飚升流行。早期发现和预防T2DM已成为全球重要公共卫生问题!目前,国外趋向于采用风险评估工具对T2DM发病风险进行预测,早期找出高危人群进行干预以控制危险因素,从而减少T2DM发生。本项目拟在前期T2DM巢式病例对照研究基础上,采用Logistic回归、聚类分析、主成分分析等数据挖掘方法,结合DELPHI专家意见筛选模型变量;采用新颖、稳健的SVM技术,以MATLAB及MATLAB SVMTOOL为平台,综合基因、环境因素及其交互作用对T2DM的影响,构建T2DM风险模型;应用ROC评估模型的灵敏度、特异度及预测精度;通过另一研究队列前瞻验证模型的外部真实性。旨为T2DM预警、预防和个性化健康服务开拓新思路,提供新措施,最终降低T2DM疾病负担,促进人群健康。
近十多年,无论是发达国家还是发展中国家,均投入大量经费用于 T2DM 的治疗,但 T2DM 的发病率、患病率和病死率仍然居高不下,危险因素仍飚升流行。早期发现和预防 T2DM 已成为全球重要公共卫生问题!目前,国外趋向于采用风险评估工具对 T2DM 发病风险进行预测,早期找出高危人群进行干预以控制危险因素,从而减少 T2DM 发生。本项目在前期 T2DM 巢式病例对照研究基础上,采用 Logistic 回归、聚类分析、主成分分析等数据挖掘技术,结合 DELPHI 专家意见筛选模型预测变量;采用神经网络和新颖、稳健的支持向量机(SVM)技术,以 MATLAB及 MATLAB SVMTOOL 为平台,综合基因、环境因素及其交互作用对 T2DM 的影响,同时考虑纳入不同变量级别,构建 T2DM风险预测模型;应用 ROC 曲线下面积评估风险预测模型的灵敏度、特异度及预测精度;通过Meta分析、T2DM 发病机制以及与 T2DM 关联结果稳定性、一致性上筛选基因;建立另一研究队列人群,前瞻性验证模型的外部真实性。主要结果:①应用两种不同建模方法,分别构建了两种不同变量级别的T2DM风险预测模型并进行了评估;②建立了12019人的模型验证队列进行基线调查和随访,补充了课题组在前一国家自然科学基金基础上建立的生物标本库中的血液样本;③检索了部分候选基因与T2DM关联性的病例-对照研究、队列研究和基础研究文献,进行了系统评价和Meta分析;④选择30个位点,对随访的新发病例和匹配的对照进行SNP检测分析,发现6个位点与研究人群T2DM发病相关;⑤风险模型应用于验证队列人群,不管是哪种建模方法,均以纳入遗传基因信息的预测模型具有较高的预测精度。这些结果可为 T2DM 预警、预防开拓新思路,为临床医生对处于不同危险度等级的个体实施不同力度的干预提供科学依据,最终控制、降低T2DM的发生,促进人类健康。
{{i.achievement_title}}
数据更新时间:2023-05-31
Efficient photocatalytic degradation of organic dyes and reaction mechanism with Ag2CO3/Bi2O2CO3 photocatalyst under visible light irradiation
粗颗粒土的静止土压力系数非线性分析与计算方法
Empagliflozin, a sodium glucose cotransporter-2 inhibitor, ameliorates peritoneal fibrosis via suppressing TGF-β/Smad signaling
基于LASSO-SVMR模型城市生活需水量的预测
基于SSVEP 直接脑控机器人方向和速度研究
基于SVM技术构建腰椎退变源性下腰痛发病风险预测模型研究
管线渗漏诱发城市地面塌陷灾变机理及其基于SVM技术风险预测模型研究
基于LASSO的乳腺癌生存者性功能障碍风险预测模型构建与验证
盐敏感性高血压候选基因的验证及其风险预测模型的构建