Research Article
Assessing Dry Weight of Hemodialysis Patients via Sparse Laplacian Regularized RVFL Neural Network with L2,1-Norm
Table 3
Bland–Altman plot analysis for different models.
| | Model | Differences with DW (%) | Limits of agreement (%) | | Mean | SD | 95% confidence interval | Lower limit | Upper limit | Number (ratio) of outside agreement interval |
| | BCM | -1.8232 | 2.7466 | -2.0706 to -1.5759 | -7.2066 | 3.5601 | 30/476 (6.30%) | | LR | 0.0002 | 2.4269 | -0.2184 to 0.2187 | -4.7566 | 4.7569 | 21/476 (4.41%) | | ANN (BP) | 0.1152 | 2.5139 | -0.1112 to 0.3416 | -4.8119 | 5.0424 | 22/476 (4.62%) | | MKRR | -0.0801 | 2.5007 | -0.3053 to 0.1451 | -4.9814 | 4.8212 | 23/476 (4.83%) | | MKSVR | -0.2638 | 2.3372 | -0.4743 to -0.05329 | -4.8446 | 4.3171 | 22/476 (4.62%) | | SLapRVFL (our method) | 0.0867 | 2.2202 | -0.1133 to 0.2866 | -4.2650 | 4.4383 | 20/476 (4.20%) |
|
|
The results are from previous work on MKSVR [ 39]. |
