Your privacy, your choice

We use essential cookies to make sure the site can function. We also use optional cookies for advertising, personalisation of content, usage analysis, and social media.

By accepting optional cookies, you consent to the processing of your personal data - including transfers to third parties. Some third parties are outside of the European Economic Area, with varying standards of data protection.

See our privacy policy for more information on the use of your personal data.

for further information and to change your choices.
Skip to main content

Table 3 Comparison of mean predictive validity of five ML models, Padua and proposed model

From: A new risk assessment model of venous thromboembolism by considering fuzzy population

Model Name

Training set

Test set

Sensitivity

Specificity

Sensitivity

Specificity

SVM

0.7894 ± 0.0220

0.7240 ± 0.0431

0.8341 ± 0.0420

0.7032 ± 0.0424

RF

0.8413 ± 0.0150

0.7757 ± 0.0138

0.8195 ± 0.0452

0.7349 ± 0.0156

GBDT

0.7883 ± 0.0404

0.8135 ± 0.0394

0.8146 ± 0.0396

0.7825 ± 0.0441

LR

0.7397 ± 0.0304

0.7960 ± 0.0230

0.8195 ± 0.0293

0.7869 ± 0.0230

XGBoost

0.7524 ± 0.0316

0.8328 ± 0.0185

0.8293 ± 0.0218

0.8064 ± 0.0239

Padua

0.8466

0.6127

0.9024

0.6330

Proposed method

0.8995 ± 1.110E-16

0.6741 ± 0.0056

0.9024 ± 1.110E-16

0.6453 ± 0.0033

  1. Values of sensitivity and specificity were represented with ‘mean value ± standard deviation’. The model training process was repeated five times to calculate the predictive validity. Note that sensitivities and specificities on training process were computed using all patients from training data
Back to article page

BMC Medical Informatics and Decision Making

ISSN: 1472-6947

Contact us