基于Logistic回归和决策树构建卵巢癌病人营养风险预测模型

打开文本图片集

AbstractObjectie：Tocostructautrioaliskpredictioodelfovariancacerpatitsasedonlisicegeionndasiicato andregressiontreCATMetos：Atotalof326varancancerpatentsadmitedtotheoncologydepartmentoftertiaryspialin Jiangxi provincefromNovember2023 to September2024 wereselectedasthestudysubjects byconveniencesampling.Generaldataof thepatientswerecolectedTeutritionaliskofthepatientswaaessed.Apredictiemodelforutritionalrskinovarancancer patientswasestablishedbasedonLogisticregresionanddecisiontre models.InfluencingfactorswereanalyzedResults：Theincidenceof nutritional risk in ovarian cancer patients was 50.92% .TheLogistic regression analysis showed that average monthly income，tumor pathological type，chemotherapy，and hypoproteinemia were influencing factorsofnutritional risk inovariancancer patients .-0.05 .The decision tree modelresultsdicatedthatemotherapy，tmorpathologicaltypeerageonthlyicome，yoalbuminemia，ag，ndplot status wereinfluencingfactorsofnutritionalriskinovariancancerpatients.Theareaunderthereceiveroperatingcharacteristiccurve （AUC）fortheLogistic modelwas 0.80.Theaccuracy was0.73.The sensitivity was 0.819.The specificitywas0.644.TheYouden's indexwasO.463.TheAUCforthedecisiontree modelwasO.802.TheaccuracywasO.742.ThesensitivitywasO.831.Thespecifity wasO.650.TheYouden'sindex wasO481.Concusios：oththeLgisicregresionanddecisiontreemodelsshowedgooddiscinatory ability.Tecombineduseoftwomodelswasbeneficialforteearlyidenticationandmanagementofnutritionaliskinovariancancer patients.Itcouldprovideaeferenceforthecomprehensiveprevention，treatment，andrehabilitationmanagementofgynecological malignant tumors.

Keywordsovarian cancer; nutritional risk; Logistic regression;decision tree; risk prediction; influencing factors摘要目的：基于Logistic回归和分类与回归树（CART）算法构建卵巢癌病人营养风险预测模型。（剩余9875字）