Skip to main content

Table 1 Characteristics of participants in the development and internal validation datasets

From: Improving accuracy of estimating glomerular filtration rate using artificial neural network: model development and validation

VariableDevelopment dataset (N = 1075)Internal validation dataset (N = 877)P value
Age (year)55.6 ± 14.557.4 ± 13.40.008
Male sex, N (%)616 (57.3)517 (59.0)0.5
Body mass index (kg/m2)24.0 ± 3.624.5 ± 3.80.005
Serum creatinine (mg/dL)*1.5 ± 1.31.3 ± 0.9< 0.001
Serum cystatin C (mg/L)1.5 ± 0.91.3 ± 0.7< 0.001
Blood urea nitrogen (mg/dL)20.9 ± 13.521.5 ± 11.70.3
Albumin (g/dL)3.9 ± 0.53.9 ± 0.50.3
Uric acid (mg/dL)6.7 ± 2.17.1 ± 2.2< 0.001
Hemoglobin (g/L)12.3 ± 2.312.7 ± 2.1< 0.001
Measured GFR (mL/min/1.73 m2)71.0 ± 27.468.8 ± 27.10.08
Measured GFR N(%)0.04
 < 30 mL/min/1.73 m270 (6.5)55 (6.3) 
 30–59 mL/min/1.73 m2310 (28.8)304 (34.7) 
 60–89 mL/min/1.73 m2420 (39.1)324 (36.9) 
 ≥ 90 mL/min/1.73 m2275 (25.6)194 (22.1) 
  1. Values for continuous variables were reported as mean ± standard deviation, and values for categorical variables as number (percentage). Conversion factor for units: serum creatinine in mg/dL to mmol/L × 88.4
  2. GFR glomerular filtration rate