Stacking classifiers based on integrated machine learning model: fusion of CT radiomics and clinical biomarkers to predict lymph node metastasis in locally advanced gastric cancer patients after neoadjuvant chemotherapy

Table 2 Diagnostic Performance Comparison of Simple versus Integrated Machine Learning Models

Model	Accuracy	AUC	F1 Score	Precision	Recall	NPV	PPV
RF	0.833	0.848	0.837	0.837	0.837	0.829	0.837
XGBoost	0.786	0.842	0.78	0.821	0.744	0.711	0.769
LightGBM	0.714	0.85	0.7	0.757	0.651	0.681	0.757
GBDT	0.714	0.808	0.733	0.702	0.767	0.73	0.702
AdaBoost	0.762	0.802	0.756	0.795	0.721	0.733	0.795
CatBoost	0.786	0.845	0.786	0.805	0.767	0.767	0.805
Hard voting	0.762	0.79	0.761	0.768	0.762	0.717	0.789
Soft voting	0.786	0.85	0.786	0.786	0.786	0.78	0.791
Stacking Classifier	0.786	0.859	0.786	0.805	0.767	0.786	0.81

GBDT gradient boosting decision tree; AdaBoost adaptive boosting: XGBoost extreme gradient boosting; LightGBM light gradient boosting machine; CatBoost categorical boosting: RF random forest; AUC area under the curve; NPV negative predictive value; PPV positive predictive value

ISSN: 1471-2407