Skip to main content

Table 2 Results of performance comparison of classification modules

From: Body landmarks and genetic algorithm-based approach for non-contact detection of head forward posture among Chinese adolescents: revitalizing machine learning in medicine

Modules

Accuracy (%, 95% CI)

Sensitivity (%, 95% CI)

Specificity (%, 95% CI)

PPVa (%, 95% CI)

NPVb (%, 95% CI)

XGBoost

78.2 (71.1, 84.2)

81.6 (72.7, 88.5)

72.6 (59.8, 83.1)

83.2 (74.4, 89.9)

70.3 (57.6, 81.1)

GBC c

81.2 (74.4, 86.9)

78.6 (69.5, 86.1)

85.4 (74.2, 93.1)

90.0 (81.9, 95.3)

70.7 (59.0, 80.6)

ETC d

82.4 (75.7, 87.9)

80.6 (71.6, 87.7)

85.5 (74.2, 93.1)

90.2 (82.2, 95.4)

72.6 (60.9, 82.4)

  1. a PPV denotes positive predictive value; b NPV denotes negative predictive value; c Gradient Boosting Classifier; d Extra Tree Classifier
Back to article page

BMC Medical Informatics and Decision Making

ISSN: 1472-6947

Contact us