Table 1.
Linear classification models separating male from female subjects within each diagnostic group
| Model | Stand. coeff. | Assigned male | Assigned female | Total | Accuracy | NPV | PPV | Specificity | Sensitivity | ||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Healthy Ctrl | |||||||||||
| mHC vs fHC at T0 | Il-2 T0 | −0.81 | 19 | 2 | 21 | True male | 0.87 | 0.90 | 0.78 | 0.90 | 0.78 |
| Il-23 T0 | 1.14 | 2 | 7 | 9 | True Female | ||||||
| mHC vs fHC at T0-T2 | Il-2 T1 | −0.72 | 19 | 2 | 21 | True male | 0.90 | 0.95 | 0.80 | 0.90 | 0.89 |
| Il-5 T1 | −0.59 | 1 | 8 | 9 | True Female | ||||||
| Il-23 T0 | 1.24 | ||||||||||
| CFS | |||||||||||
| mCFS vs fCFS at T0 | Il-8 T0 | 1.00 | 7 | 5 | 12 | True male | 0.68 | 0.78 | 0.62 | 0.58 | 0.80 |
| 2 | 8 | 10 | True Female | ||||||||
| mCFS vs fCFS at T0-T2 | TNFb T2 | −0.91 | 8 | 4 | 12 | True male | 0.77 | 0.89 | 0.69 | 0.67 | 0.90 |
| Il-6 T1 | −0.83 | 1 | 9 | 10 | True Female | ||||||
| Il-8 T0 | 1.75 | ||||||||||
| GWI | |||||||||||
| mGWI vs fGWI at T0 | NA | NA | NA | NA | NA | NA | |||||
| mGWI vs fGWI at T0-T2 | Il-2 T2 | −1.51 | 19 | 1 | 20 | True male | 0.87 | 0.86 | 0.88 | 0.95 | 0.70 |
| IL-1a T0 | 0.60 | 3 | 7 | 10 | True Female | ||||||
| IL-1a T1 | −1.08 | ||||||||||
| Il-6 T1 | 0.84 | ||||||||||
| Il-10 T1 | 1.36 |
Performance values correspond to leave-one-out cross validation results.