Results of ROC Curve Analysis for Detecting Liver Cirrhosis
| Parameter | Liver | Spleen | ||||
|---|---|---|---|---|---|---|
| Volume (cm3) | %ID/cm3 | Uptake (%) | Volume (cm3) | %ID/cm3 | Uptake (%) | |
| Sensitivity (%) | 38 (26/52) | 70 (57/81) | 78 (66/88) | 90 (79/96) | 33 (22/47) | 78 (66/88) |
| Specificity (%) | 78 (61/90) | 94 (81/99) | 97 (85/99) | 100 (100/100) | 83 (67/93) | 100 (100/100) |
| PPV (%) | 74 (62/85) | 95 (86/99) | 98 (91/100) | 100 (100/100) | 77 (64/87) | 100 (100/100) |
| NPV (%) | 43 (27/61) | 65 (49/81) | 73 (55/86) | 86 (70/95) | 43 (26/60) | 73 (56/87) |
| AUC | 0.51 (0.41/0.62) | 0.89 (0.81/0.94) | 0.94 (0.87/0.94) | 0.97 (0.91/0.99) | 0.52 (0.41/0.62) | 0.93 (0.86/0.97) |
| Threshold | 1,563 | 0.028 | 47.6 | 389 | 0.026 | 14.9 |
PPV and NPV = positive and negative predictive values, respectively; AUC = area under curve.
Numbers in parentheses are 95% confidence intervals.