Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

< Back to Article

Table 1.

Comparison of baseline characteristics between the LAT/SEC and control cohorts.

More »

Table 1 Expand

Fig 1.

LASSO regression analysis.

(a) The use of 10-fold cross-validation to draw vertical lines at selected values, where the optimal lambda produces 5 nonzero coefficients. (b) In the LASSO model, the coefficient profiles of 23 texture features were drawn from the log (λ) sequence. Vertical dotted lines are drawn at the minimum mean square error (λ = 0.007) and the standard error of the minimum distance (λ = 0.026).

More »

Fig 1 Expand

Table 2.

Multivariate logistic regression analysis.

More »

Table 2 Expand

Fig 2.

Comprehensive analysis of ML models.

(a) ROC curve analysis of ML algorithms for the prediction of LAT/SEC in the training set. (b) ROC curve analysis of ML algorithms for the prediction of LAT/SEC in the validation set. (c) Calibration plots for predicting LAT/SEC in NVAF patients using various models. (d) Forest Plot of each model AUC score.

More »

Fig 2 Expand

Table 3.

Predictive performance of 9 ML algorithms in training and validation sets for LAT/SEC in NVAF patients.

More »

Table 3 Expand

Fig 3.

Logistic regression model evaluation.

(a-c) The ROC curves of logistic regression using the 10-fold cross-validation on the training set (a), validation set (b), and test set (c). (d) Machine learning curve. (e) calibration plots for logistic regression. (f) Decision curve analysis graph showing the net benefit against threshold probabilities based on decisions from model outputs.

More »

Fig 3 Expand

Table 4.

Diagnostic performance of the logistic regression model for the prediction of LAT/SEC risk in NVAF.

More »

Table 4 Expand

Fig 4.

SHAP analysis of the model.

(a) Feature attributions in SHAP. Each line corresponds to a feature, with SHAP values plotted on the abscissa. Red dots denote higher values, while blue dots represent lower values. (b) Importance of variables depicted as bars, indicating their contribution to model predictions. (c) SHAP scores elucidate the predicted risk of LAT/SEC in an individual subject.

More »

Fig 4 Expand

Fig 5.

ROC of the logistic regression model and CHA2DS2-VASc in predicting LAT/SEC risk in NVAF.

More »

Fig 5 Expand

Table 5.

Performance comparison of the proposed ML-based logistic regression model with CHA2DS2-VASc in predicting LAT/SEC risk in NVAF.

More »

Table 5 Expand