Skip to main content

Table 7 Improved predictive performance using only selected features in decision trees, random forests, and support vector machines for healthy controls (HCs) versus rheumatoid arthritis (RA) patients and osteoarthritis (OA) versus RA patients

From: Isotypes of autoantibodies against novel differential 4-hydroxy-2-nonenal-modified peptide adducts in serum is associated with rheumatoid arthritis in Taiwanese women

Algorithm* Accuracy Precision F1 score Sensitivity Specificity AUC Selected feature
HC v.s. RA
 DT 0.84 (0.80–0.87) 0.92 (0.88–0.95) 0.80 (0.76–0.85) 0.73 (0.67–0.80) 0.93 (0.89–0.97) 0.86 (0.82–0.89) IgM anti-HPT78−108 HNE
IgG anti-THRB328−345
 RF 0.87 (0.85–0.90) 0.91 (0.87–0.95) 0.85 (0.82–0.88) 0.81 (0.76–0.87) 0.93 (0.91–0.96) 0.92 (0.90–0.95) IgM anti-HPT78−108 HNE
IgM anti-HPT78−108
IgG anti-HPT78−108
IgG anti-THRB328−345
HNE adducts
 SVM 0.71 (0.62–0.80) 0.97 (0.91–1.00) 0.56 (0.42–0.70) 0.42 (0.27–0.58) 0.98 (0.95–1.00) 0.82 (0.75–0.89) IgM anti-HPT78−108 HNE
IgG anti-HPT78−108 HNE
IgM anti IGKC2−19
IgM anti-CFAH1121−1230 HNE
OA v.s. RA
 DT 0.81 (0.75–0.87) 0.94 (0.89–0.98) 0.82 (0.76–0.89) 0.74 (0.62–0.83) 0.93 (0.88–0.98) 0.84 (0.79–0.89) IgM anti-HPT78−108 HNE IgG anti-THRB328−345 IgG anti-IGKC2−19
 RF 0.85 (0.81–0.89) 0.87 (0.82–0.91) 0.87 (0.82–0.92) 0.88 (0.79–0.97) 0.79 (0.71–0.87) 0.92 (0.88–0.96) IgM anti-HPT78−108 HNE IgG anti-HPT78−108 HNE IgG anti-IGKC2−19 IgM anti-CFAH1121−1230 HNE
 SVM 0.68 (0.61–0.76) 0.68 (0.61–0.76) 0.81 (0.75–0.86) 0.53 (0.31–0.66) 0.94 (0.89–0.98) 0.88 (0.83–0.93) IgM anti-HPT78−108 HNE IgG anti-THRB328−345
  1. *DT, decision trees; RF, random forests; SVM, support vector machines
\