Table 3.

Sample classification performance

Set 1Set 2
Healthy vs. cancerHealthy vs. cancer
MethodSensitivitySpecificityErrorNo. antibodiesMethodSensitivitySpecificityErrorNo. antibodies
10.8130.9200.1392510.8630.8950.12114
20.8800.8950.1132820.9500.8950.07420
30.9000.8960.1021030.9330.9380.0656
Healthy vs benignHealthy vs benign
MethodSensitivitySpecificityErrorNo. antibodiesMethodSensitivitySpecificityErrorNo. antibodies
10.8420.8900.129610.7170.8950.17814
20.8080.9200.1271320.7580.9400.13517
30.9671.0000.013630.9700.9790.0258
Cancer vs benignCancer vs benign
MethodSensitivitySpecificityErrorNo. antibodiesMethodSensitivitySpecificityErrorNo. antibodies
10.6500.6000.366210.5430.9000.3321
20.8830.4830.258420.8300.4830.2902
30.9490.8000.1011130.9160.7420.1458
  • NOTE: Method 1, boosting decision tree; method 2, boosting logistic regression; and method 3, logistic regression with forward selection. The average sensitivities, specificities, and error rates from the cross-validation analyses, as well as the number of antibodies used in the classifiers, are indicated.