BMC Medical Informatics and Decision Making

Table 3 Results of the model performance on training and test sets

From: Development of machine learning models for detection of vision threatening Behçet’s disease (BD) using Egyptian College of Rheumatology (ECR)–BD cohort

	Training set (N = 840)						Test set (N = 209)
	AUROC (95%CI)	Accuracy	Sensitivity	Specificity	PPV	NPV	AUROC (95%CI)	Accuracy	Sensitivity	Specificity	PPV	NPV
XGBoost	0.98 (0.97,0.99)	0.79	0.99	0.98	0.97	0.99	0.85 (0.81,0.90)	0.85	0.85	0.86	0.86	0.84
RF	0.99 (0.98,0.99)	0.99	0.99	0.99	0.99	0.99	0.83 (0.79,0.89)	0.83	0.83	0.84	0.84	0.82
Extra tree	0.99 (0.98,0.99)	0.99	0.98	0.99	0.99	0.98	0.79 (0.69,0.89)	0.79	0.81	0.77	0.79	0.80
SVM	0.72 (0.69,0.75)	0.73	0.66	0.84	0.89	056	0.78 (0.73,0.83)	0.78	0.73	0.86	0.90	0.66
ANN	0.66 (0.62,0.69)	0.57	0.64	0.56	0.35	0.81	0.67 (0.61,0.73)	0.67	0.68	0.66	0.66	0.67
MLP	0.72 (0.67,0.78)	0.70	0.71	0.69	0.67	0.73	0.66 (0.59,0.72)	0.66	0.71	0.63	0.57	0.76
LR	0.67 (0.63,0.69)	0.67	0.72	0.63	0.55	0.78	0.64 (0.58,0.71)	0.65	0.75	0.60	0.46	0.84

Models are listed in order of decreasing AUROC
AUROC, area under the curve; XGBoost, extreme gradient boosting; RF, random forest; SVM, support vector machine; ANN, artificial neural networks; MLP, multi-layer perceptron; LR, logistic regression; PPV, positive predictive value; NPV, negative predictive value

Back to article page

ISSN: 1472-6947

Contact us

General enquiries: journalsubmissions@springernature.com