基于可解释机器学习算法构建乳腺癌病人术前衰弱风险预测模型

打开文本图片集

AbstractObjective：Toanalyetheinfluencingfactorsofpreoperativefrailtyinbreastcancerpatientsandtodevelopariskprediction model.Methds：Atotalof583inpatientsscheduledforsurgical treatmentinthebreastsurgerydepartmentsof twoGradeAtertary hospitals in Binzhoucity，Shandong province，were selected betwenAugust 2O24andJanuary 2O25.Theywererandomlydivided intoa training set（4O8 cases） and a validation set（175 cases） in a 7：3 ratio.Variables were screened using both Lasso regression and the Boruta algorithm.Fourmacineleaingalgoritssupportvectormachin，ecisiontre，ligtgadietboostingmachne，ndxtremegadit boosting-wereemployedtodeveloppredictionmodels.Modelperformancewascomparedbasedonaccuracy，precisionsensitivity， specfity，F1-score，andthareaunder tereceiveroperatingcharacteristic（OC）urve（AUC）.Andtheoptialmodelwasintepeted using the SHAP method.Results： The incidence of preoperative frailty in breast cancer patients was 26.93% .Compared to the support vectormachine，decisiontree，andligtgradientbostingmachinemodels，theextremegradientboostingmodeldemonstratedtebest performance，withanAUCof0.99，accuracyof.829，precisionofO646，snsitivityof0.857spcifcityofO.818，andanFcoreof 0.737.The SHAPbarplotidentifiedthetopfiveinfluencingfactorsasage，hemoglobin，albumin，neutrophilpercentage，and comorbidities.Conclusios：Theextremegradientboostingmodelexibits thebestpredictiveperformanceandcanserveasareliabletool for healthcare providers to efectively assess and scientifically manage preoperative frailty in breast cancer patients.

Keywordsbreastcancer;frailty;preoperative;machine learning;prediction model;influencing factors;comorbidities乳腺癌是全球女性发病率最高的恶性肿瘤之一，也是女性癌症相关死亡最主要的原因[1]。（剩余12599字）