Accurate Prediction of Advanced Liver Fibrosis Using the Decision Tree Learning Algorithm in Chronic Hepatitis C Egyptian Patients

Table 2

Characteristics of variables in coherent dataset.

Characteristics	Training dataset 22690	Validation dataset 16877	Pearson correlation coefficients	value
Age (yrs)	40 ± 11	40 ± 10	0.26	<0.0001
Gender
Female	6186 (27.3%)	4555 (26.9%)	−0.03	0.008
Male	16504 (72.7%)	12322 (73.1%)
BMI	26.70 ± 3.79	26.79 ± 3.84	0.10	<0.0001
AFP (U/L)	7.26 ± 26.61	7.69 ± 28.49	0.10	<0.0001
ALP (U/L)	105.41 ± 65.17	105.41 ± 65.17	0.02	0.008
AST (U/L)	57.27 ± 33.73	56.78 ± 34.61	0.12	<0.0001
ALT (U/L)	61.84 ± 36.89	61.84 ± 38.19	0.06	0.008
Platelet count (10^9/L)	212.48 ± 60.64	211.55 ± 60.86	−0.18	<0.0001
Albumin (g/dL)	4.39 ± 0.42	4.40 ± 0.42	−0.14	<0.0001
Indirect bilirubin (mg/dL)	0.57 ± 1.77	0.60 ± 2.24	−0.00	0.088
Total bilirubin (mg/dL)	0.76 ± 0.28	0.76 ± 0.28	0.05	<0.0001
Glucose (mg/dL)	96.57 ± 19.41	96.69 ± 20.71	0.08	<0.0001
Hemoglobin (Hb)	14.03 ± 1.47	14.03 ± 1.62	−0.00	0.0005
WBC (10^9/L)	6.44 ± 1.90	6.44 ± 1.94	−0.02	0.0001