4.

基于朴素贝叶斯分类器方法的CINV风险预测模型与其他研究结果对比

Comparison of CINV prediction model based on machine learning method with other research models

Model	AUC	Sensitivity	Specificity
Prediction model of acute CINV based on machine learning	0.72±0.04 (95%CI: 0.69-0.75)	0.83±0.04 (95%CI: 0.80-0.86)	0.45±0.03 (95%CI: 0.42-0.47)
Prediction model of delay CINV based on machine learning	0.74±0.02 (95%CI: 0.72-0.77)	0.84±0.01 (95%CI: 0.83-0.86)	0.48±0.03 (95%CI: 0.45-0.52)
Prediction model of acute CINV from George team	0.69 (95%CI: 0.59-0.79)	0.72	0.52
Prediction model of delay CINV from George team	0.70 (95%CI: 0.60-0.80)	0.76	0.50
Prediction model of CINV from Molassiotis team	-	0.79	0.50