Skip to main content

Table 6 Performance measurements for different experiments

From: DeepVAQ : an adaptive deep learning for prediction of vascular access quality in hemodialysis patients

Experiment

\({\mathbf{e}\mathbf{x}\mathbf{p}}_{({\varvec{i}},{\varvec{j}})}\)

Accuracy

Sensitivity

Specificity

Precision

F-Score

AUC

Loss validation

\({{\text{exp}}}_{(\mathrm{8,3})}\)

0.9021 +—0.09

0.7558 +—0.20

0.9388 +—0.07

0.7698 +—0.24

0.7610 +—0.22

0.8473 +—0.13

1.2521 +—1.02

\({{\text{exp}}}_{(\mathrm{8,5})}\)

0.8915 +—0.11

0.7267 +—0.37

0.9320 +—0.13

0.7391 +—0.28

0.6996 +—0.35

0.8294 +—0.18

1.4008 +—1.90

\({{\text{exp}}}_{(\mathrm{8,7})}\)

0.8851 +—0.12

0.7111 +—0.36

0.9280 +—0.15

0.7957 +—0.29

0.6939 +—0.33

0.8196 +—0.18

1.4810 +—1.85

\({{\text{exp}}}_{(\mathrm{16,3})}\)

0.8745 +—0.07

0.6863 +—0.27

0.9215 +—0.03

0.6681 +—0.17

0.6704 +—0.22

0.8039 +—0.14

1.6079 +—1.40

\({{\text{exp}}}_{(\mathrm{16,5})}\)a

0.9106 +—0.08a

0.7768 +—0.17a

0.9441 +—0.07a

0.7984 +—0.22a

0.7829 +—0.18a

0.8605 +—0.11a

1.1441 +—0.86a

\({{\text{exp}}}_{(\mathrm{16,7})}\)

0.9000 +—0.10

0.7512 +—0.25

0.9376 +—0.08

0.7655 +—0.25

0.7490 +—0.25

0.8444 +—0.15

1.2754 +—1.30

\({{\text{exp}}}_{(\mathrm{32,3})}\)

0.8830 +—0.11

0.7077 +—0.23

0.9269 +—0.07

0.7287 +—0.28

0.7169 +—0.26

0.8173 +—0.16

1.4985 +—1.19

\({{\text{exp}}}_{(\mathrm{32,5})}\)

0.8915 +—0.11

0.7300 +—0.25

0.9323 +—0.09

0.7582 +—0.29

0.7335 +—0.26

0.8311 +—0.15

1.3840 +—1.30

\({{\text{exp}}}_{(\mathrm{32,7})}\)

0.9021 +—0.09

0.7560 +—0.19

0.9388 +—0.07

0.7794 +—0.25

0.7635 +—0.21

0.8474 +—0.13

1.2506 +—0.98

  1. a Best classification result