Table 2 Equation performance in validation dataset (n = 266).

From: Deep learning body-composition analysis of clinically acquired CT-scans estimates creatinine excretion with high accuracy in patients and healthy individuals

Formula

Bias

Precision

Accuracy

MPE [95% CI]

R2 [95% CI]

RMSE [95% CI]

MAPE [95% CI]

P15 [95% CI]

P30 [95% CI]

Cockcroft–Gault

− 0.16 [− 0.60; 0.27]

0.40 [0.28; 0.52]

3.52 [3.07; 3.96]

27.2% [22.7–31.7%]

40% [34%; 46%]

77% [72%; 82%]

Ix

0.84 [0.44; 1.25]

0.42 [0.31; 0.52]

3.46 [3.07; 3.85]

30.3% [25.2–35.5%]

46% [40%; 52%]

71% [66%; 77%]

CRAFT 1

0.18 [− 0.13; 0.50]

0.63 [0.53; 0.72]

2.68 [2.34; 3.01]

21.0% [17.7–24.2%]

56% [50%; 62%]

81% [76%; 85%]

CRAFT 2

0.16 [− 0.17; 0.48]

0.61 [0.50; 0.70]

2.78 [2.43; 3.12]

22.3% [18.8–25.8]

49% [44%; 55%]

80% [76%; 85%]

  1. Confidence intervals were calculated with the combined variance of multiple imputation (10×) and bootstrap (1000×).
  2. MPE mean prediction error (mmol/day), MAPE mean absolute percentage error, RMSE root mean squared error (mmol/day), R2 the R2-value calculated with linear regression, p15/p30 the percentage of points that fall within 15%/30% of the outcome.
Back to article page