老年结直肠癌患者长期随访发生心力衰竭风险的人工智能模型分析  

Evaluation of the risk of tumor-related functional insufficiency in aging colorectal cancer patients during treatment

作　　者：戴荣 张爱华 曹荣元[2] DAI Rong;ZHANG Aihua;CAO Rongyuan(Department of Oncology,The Second People’s Hospital of Lianyungang,Lianyungang 222002,China)

机构地区：[1]连云港市第二人民医院肿瘤科,222002 [2]连云港市第二人民医院心内科,222002

出　　处：《心肺血管病杂志》2024年第3期255-260,共6页Journal of Cardiovascular and Pulmonary Diseases

摘　　要：目的:应用人工智能(artificial intelligence,AI)方法比较三种模型预测老年结直肠癌患者长期随访发生相关心力衰竭(cancer therapy-related cardiac dysfunction,CTRCD)风险的临床价值。方法:单中心回顾性队列研究。选取2010年1月至2018年12月,在连云港市第二人民医院住院诊治的老年结直肠癌患者382例。按照2:1随机分为训练集(255例)和验证集(127例)。比较两组基线和预后资料。随访2年观察终点为:CTRCD。应用深度学习方法分别建立和验证预测CTRCD发生风险的三种模型,包括逻辑回归模型、随机森林模型(random forest model,RF)、极端梯度提升模型(extreme gradient boosting,XG Boost)。比较三种模型对CTRCD的预测效能,比较三种模型预测CTRCD的准确性和有效性。结果:CTRCD的发生率为15.2%(58例),其中训练集和验证集分别有37例(14.5%)和21例(16.5%)。与训练集比较,验证集患者的吸烟比例更高(P<0.05),其他基线资料均相似。在训练集中,三种模型预测CTRCD的AUC分别为0.75、0.79和0.84,但逻辑回归模型的特异性很低(0.56),准确率也不高(0.66),而RF模型的预测准确率也低于XGBoost模型(0.74 vs.0.81)。在验证集中,三种模型预测CTRCD的AUC分别为0.72、0.77和0.82。同时,对于XGBoost模型,训练集和验证集的校准曲线显示,对于XGBoost模型预测CTRCD的实际观察结果与列线图预测结果之间有很好的相关性。训练集和验证集的临床决策曲线均显示,XGBoost模型的净获益值较高,有效性较好。结论:XGBoost模型预测结直肠癌患者发生CRTCD的准确性和稳定性最优,重要相关因素包括整体纵向应变、LVEF、cTnI、利钠肽和年龄等。Objective:To compare three models for predicting the risk of tumor treatment-related cardiac dysfunction(CTRCD)in aging colorectal cancer patients using deep learning methods.Methods:This was a retrospective cohort study.From January 2010 to December 2018,there were 382 aging patients with colorectal cancer who were hospitalized and treated at the Second People's Hospital of Lianyungang City.All those patients were randomly divided into the training group(255 cases)and the validation group(127 cases)based on a 2:1 ratio.The baseline and prognostic data between two groups were compared.All patients were followed up for at least 2 years.The primary endpoint was CTRCD.The deep learning method was used to establish and verify three models to predict the risk of CTRCD,including logical regression model,random forest model(RF),and extreme gradient lifting model(XG Boost).We used the AUC and ROC to compare the predictive performance of three models for CTRCD,and compare the accuracy and effectiveness in predicting CTRCD using calibration curves and clinical decision curves.Results:Among the 382 patients,the incidence of CTRCD was 15.2%(58 cases),with 37 cases(14.5%)in the training group and 21 cases(16.5%)in the validation group,respectively.Compared with the training group,the smoking proportion of patients in the validation group was higher(P<0.05),and other baseline data were similar between the two groups.In the training group,the AUC of the logistic regression model,RF,and XG Boost models for predicting CTRCD were 0.75,0.79,and 0.84,respectively.However,the specificity and accuracy of the logistic regression model were very low(0.56)and not high(0.66),and the prediction accuracy of the RF model was also lower than that of the XGBBoost model(0.74 vs.0.81).In the validation group,the AUC predicted by the three models for CTRCD were 0.72,0.77,and 0.82,respectively.Meanwhile,the calibration curves in the training group and validation group showed good correlation between the actual observation results of the XGBoost mo

关 键 词：直肠癌 肿瘤相关心力衰竭 深度学习 预测价值 

分 类 号：R54[医药卫生—心血管疾病]