Skip to main content

Table 4 Blind Test Performance for the 33, 21, and LCDT1 Algorithm (Validation Set)

From: Blood test shows high accuracy in detecting stage I non-small cell lung cancer

 Model
StatisticsAlgorithm 33Algorithm 21LCDT1 Algorithm
Accuracy95.6% (92.4, 97.7%)94.3% (90.7, 96.8%)95.6% (92.4, 97.7%)
Sensitivity89.1% (78.9, 95.3%)89.1% (78.9, 95.3%)89.1% (78.9, 95.3%)
Specificity97.7% (94.6, 99.2%)96.0% (92.2–98.2%)97.7% (94.6, 99.2%)
Positive Predictive Value (PPV)92.5% (83.0, 97.4%)87.5% (77.0, 94.2%)92.5% (83.0, 97.4%)
Negative Predictive Value (NPV)96.6% (93.1, 98.6%)96.5% (93.0, 98.5%)96.6% (93.1, 98.6%)
NSCLC Prevalence24.1%24.1%24.1%
True Positive (TP)494449
True Negative (TN)169166169
False Positive (FP)464
False Negative (FN)676
  1. All entries show the statistical (95% CI). *Other cancer types were included in the analysis. Each subject consisted of two replicates (N = 2) or two data points processed by the algorithm. If one data point was positive, then the subject was considered positive for LC. Table was generated using R Version 3.4.4