A new prediction model for sustained ventricular tachycardia in arrhythmogenic cardiomyopathy

Figures

• 

  Figure 1. The least absolute shrinkage and selection operator (LASSO) regression model was employed to identify factors associated with sVT in ACM patients. (A) The optimal λ selection in the LASSO model was determined using 10-fold cross-validation, guided by the minimum criteria (the right dotted vertical line) and the 1-SE criteria (the left dotted vertical line). (B) The LASSO coefficient profiles for all 23 features revealed that eleven features with nonzero coefficients were selected at the optimal λ, as indicated by the vertical line.

• 

  Figure 2. The nomogram was constructed in the training group for identifying ACM patients at high risk of sVT. Each factor in the nomogram was assigned an individual score based on its value, and the scores of all factors were summarized to a total score, which was used to estimate the risk of sVT for a given patient.

• 

  Figure 3. The nomogram was validated in both the training and validation groups. (A) The ROC curve was utilized to evaluate the performance of the nomogram in the training group. The AUC of the nomogram for predicting the occurrence of sVT in ACM patients was 0.867 (95% CI, 0.797–0.938). (B) The calibration curve in the training group demonstrated that the predictive nomogram provided excellent estimations of actual probabilities of sVT. (C) The ROC curve was used to evaluate the performance of the nomogram in the validation group. The AUC of the nomogram for predicting the occurrence of sVT in ACM patients was 0.815 (95% CI, 0.673–0.958). (D) The calibration curve in the validation group validated that the predictive nomogram provided excellent estimations of actual probabilities of sVT.

• 

  Figure 4. DCA was performed for the nomogram in both the training group (A) and the validation group (B) the red line indicated the predictive nomogram. The gray line represented the assumption that sVT occurred in all ACM patients, while the black line represented the assumption that no patient had sVT.