Skip to main content

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

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

  Proportion Positive Sensitivity Specificity Positive Predictive Value Negative Predictive Value True Positive False Positive True Negative False Negative
GBMb 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
  1. Abbreviation: GBM one implementation of gradient boosting machine
  2. a Observed bloodstream infection rate 139/2236 (6.2%) in the test set
  3. 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)
\