Table 2 Classification results for osteoarthritis severity of each knee compartment using support vector machine classifier with linear kernel.

From: Using magnetic resonance imaging-based subregional texture analysis models to classify knee osteoarthritis severity by compartment

 

Accuracy

Sensitivity

Specificity

Precision

F-measure

ROC AUC

Medial compartment

 FS-PDWI

0.734

0.544

0.785

0.544

0.527

0.721

 PDWI

0.757

0.573

0.799

0.555

0.542

0.725

 Combination of FS-PDWI and PDWI

0.787

0.637

0.833

0.641

0.636

0.786

Lateral compartment

 FS-PDWI

0.622

0.635

0.608

0.622

0.631

0.608

 PDWI

0.705

0.718

0.691

0.706

0.712

0.751

 Combination of FS-PDWI and PDWI

0.694

0.676

0.712

0.694

0.692

0.758

P-FT compartment

 FS-PDWI

0.640

0.597

0.684

0.642

0.627

0.658

 PDWI

0.662

0.635

0.690

0.666

0.656

0.662

 Combination of FS-PDWI and PDWI

0.676

0.629

0.723

0.677

0.663

0.679

  1. FS-PDWI fat-suppressed proton density-weighted imaging, PDWI proton density-weighted imaging, P-FT articulation between patella and femoral trochlea.