Additional file 1 of An autopsy-based cardiac lesion evaluation system facilitates quantitative diagnosis of sudden cardiac death: development and multicenter validation of a machine learning model

Le résumé fourni par la source

Additional file 1. Supplementary methods. Table S1. Demographic and heart examination characteristics of study cohorts in five external centers. Table S2. Statistical metrics for each machine learning model in the six datasets. Table S3. Statistical metrics of nomogram corresponding to various cutoff value in the six datasets. Table S4. Statistical metrics according to the human-machine comparison and fusion experiment. Table S5. Characteristics of patients in clinical cohort. Fig. S1. Missingness map of datasets of six forensic centers. Fig. S2. The correlation heatmap of variables in the interaction of LASSO and RF-RFE method. Fig. S3. ROC curves of eight ML models in four external datasets. Fig. S4. Sensitivity analysis about the missingness during the model construction. Fig. S5. Subgroup analyses of the nomogram in natural death and sudden death mode. Fig. S6. Subgroup analyses of the nomogram in different diseases. Fig. S7. Feature selection for the prediction of sudden coronary artery death among individuals with coronary artery disease in forensic setting. Fig. S8. The two-dimensional echocardiography-available measurement of heart morphological features

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.