Development and utility assessment of a machine learning bloodstream infection classifier in pediatric patients receiving cancer treatments

Table 5 Performance in Test Set (N = 2236) ^a

	Proportion Positive	Sensitivity	Specificity	Positive Predictive Value	Negative Predictive Value	True Positive	False Positive	True Negative	False Negative
GBM^b	595 (26.6%)	0.63	0.76	0.15	0.97	87	508	1589	52
Neutropenia	679 (30.4%)	0.63	0.72	0.13	0.97	87	592	1505	52

Abbreviation: GBM one implementation of gradient boosting machine
^a Observed bloodstream infection rate 139/2236 (6.2%) in the test set
^b Test threshold chosen (probability> 0. 0489) to keep the number of false negatives the same as in the absolute neutrophil count < 0.5 × 10⁹/L model (52/2236)

ISSN: 1471-2407