Journal: Journal of Cardiothoracic Surgery

Article Title: Diagnostic value of echocardiography combined with serum C-reactive protein level in chronic heart failure

doi: 10.1186/s13019-023-02176-7

Figure Lengend Snippet: Correlation analysis of EQPs (LVEF, FS and E/A) and serum CRP level in CHF patients. Pearson method was used to analyze the correlation between LVEF and serum CRP level ( A ), FS and serum CRP level ( B ), and E/A and serum CRP level ( C ) in CHF patients

Article Snippet: MedCalc analysis illustrated that the area under the curve (AUC) of EQPs combined with serum CRP level in identifying CHF was significantly higher than that of EQPs (LVEF, FS, and E/A) or serum CRP alone (Fig. , all P < 0.05).

Techniques:

Journal: Journal of Cardiothoracic Surgery

Article Title: Diagnostic value of echocardiography combined with serum C-reactive protein level in chronic heart failure

doi: 10.1186/s13019-023-02176-7

Figure Lengend Snippet: Diagnostic efficacy of EQPs combined with serum CRP on CHF. ROC curve was used to analyze the diagnostic efficacy of the EQPs (LVEF, FS and E/A), serum CRP levels, and their combination for CHF patients

Article Snippet: MedCalc analysis illustrated that the area under the curve (AUC) of EQPs combined with serum CRP level in identifying CHF was significantly higher than that of EQPs (LVEF, FS, and E/A) or serum CRP alone (Fig. , all P < 0.05).

Techniques: Diagnostic Assay

Journal: Scientific Reports

Article Title: Neural networks versus Logistic regression for 30 days all-cause readmission prediction

doi: 10.1038/s41598-019-45685-z

Figure Lengend Snippet: Performance analysis of the tested models. Panels A–E report the average ROC curve of the best models. The optimal cutoff is based on the average Youden-Index of each model for all 5-folds. Standard deviation of the optimal cutoff position is reported on the graph. Panel F reports the cumulative average AUC performance as a function of patients’ timeline length.

Article Snippet: We used the area under the ROC curve (AUC) as our performance measure with confidence intervals computed using the approach reported in LeDell et al . .

Techniques: Standard Deviation

Journal: Scientific Reports

Article Title: Neural networks versus Logistic regression for 30 days all-cause readmission prediction

doi: 10.1038/s41598-019-45685-z

Figure Lengend Snippet: Trained models’ performance based on the area under the ROC curve (AUC). CI: confidence interval.

Article Snippet: We used the area under the ROC curve (AUC) as our performance measure with confidence intervals computed using the approach reported in LeDell et al . .

Techniques: