Research Article

A Self-Representation-Based Fuzzy SVM Model for Predicting Vascular Calcification of Hemodialysis Patients

Table 1

The information of dataset.

No.FeatureValue

1Gender (males/females)32/27-0.0455
2Age (years)0.4010
3Smoking (yes/no)1/58-0.0847
4BMI (kg/m2)0.1639
5DM (yes/no)24/350.4847
6CI (yes/no)3/560.0025
7CHD (yes/no)5/540.0433
8Systolic blood pressure (mmHg)-0.1150
9Diastolic blood pressure (mmHg)-0.2043
10Phosphate binder (yes/no)36/23-0.2141
11Hemoglobin (g/L)-0.2584
12C-reactive protein (mg/L)0.3016
13Serum creatinine (μmol/L)-0.4252
14Serum glucose (mmol/L)0.2608
15Serum calcium (mmol/L)0.0520
16Serum phosphorus (mmol/L)-0.0862
17Total glyceride (mmol/L)-0.0542
18Total cholesterol (mmol/L)-0.0466
19Low-density lipoprotein-C (mmol/L)-0.0252
20High-density lipoprotein-C (mmol/L)0.0866
21HbA1c (%)0.2151
22Serum albumin (g/L)-0.1308
2325-OH vitamin D3 (ng/mL)0.3850
24iPTH (pg/mL)-0.0225
25GNRI-0.0078
26FGF-23 (pg/mL)-0.0966
27Klotho (ng/mL)0.0443
28Interleukin-6 (pg/mL)0.2634
29Fetuin-A (pg/mL)-0.0234

Denotes that each feature correlated with ascular calcification level using the Pearson correlation coefficient ().