An Efficacy Predictive Method for Diabetic Ulcers Based on Higher-Order Markov Chain-Set Pair Analysis
Table 8
The connection degree (CD) of predicted and actual and corresponding efficacy scores (U).
Patient
Results
The five-element connection degree
Efficacy scores (U)
Relative error (%)
Patient a
Actual results
(t4) = 0.39 + 0.61i
7.78
Predicted results of higher-order Markov chain-SPA
(t4) = 0.44 + 0.38i + 0.11j + 0.08k
7.34
5.99
Predicted results of traditional Markov chain-SPA
(t4) = 0.38 + 0.36i + 0.05j + 0.02k + 0.19l
6.44
17.22
Patient b
Actual results
(t4) = 0.26 + 0.74i
7.52
Predicted results of higher-order Markov chain-SPA
(t4) = 0.30 + 0.46i + 0.19j + 0.04k
7.03
6.97
Predicted results of traditional Markov chain-SPA
(t4) = 0.45 + 0.33i + 0.07j + 0.01k + 0.14l
6.91
8.11
Patient c
Actual results
(t4) = 0.30 + 0.62i + 0.08j
7.44
Predicted results of higher-order Markov chain-SPA
(t4) = 0.30 + 0.42i + 0.25j + 0.03k
6.96
6.90
Predicted results of traditional Markov chain-SPA
(t4) = 0.45 + 0.33i + 0.07j +0 .01k + 0.14l
6.88
7.53
Patient d
Actual results
(t4) = 0.39 + 0.53i + 0.08j
7.62
Predicted results of higher-order Markov chain-SPA
(t4) = 0.39 + 0.39i + 0.15j + 0.07k
7.20
5.83
Predicted results of traditional Markov chain-spa
(t4) = 0.38 + 0.39i + 0.04j + 0.02k + 0.17l
6.58
13.65
Patient e
Actual results
(t4) = 0.26 + 0.74i
7.52
Predicted results of higher-order Markov chain-SPA
(t4) = 0.36 + 0.34i + 0.26j + 0.04k
7.06
6.52
Predicted results of traditional Markov chain-SPA
(t4) = 0.42 + 0.34i + 0.09j + 0.01k + 0.14l
6.78
9.84
Patient f
Actual results
(t4) = 0.30 + 0.47i + 0.23j
7.14
Predicted results of higher-order Markov chain-SPA
(t4) = 0.28 + 0.40i + 0.25j + 0.07k
6.78
5.15
Predicted results of traditional Markov chain-SPA
(t4) = 0.32 + 0.40i + 0.12j + 0.02k + 0.19l
6.53
8.54
Patient g
Actual results
(t4) = 0.28 + 0.31i + 0.3j + 0.11k
6.52
Predicted results of higher-order Markov chain-SPA
(t4) = 0.24 + 0.25i + 0.37j + v0.14k
6.17
5.33
Predicted results of traditional Markov chain-SPA
(t4) = 0.29 + 0.19i + 0.34j + 0.08k + 0.11l
5.97
8.50
Patient h
Actual results
(t4) = 0.62 + 0.11i + 0.19j + 0.08k
7.54
Predicted results of higher-order Markov chain-SPA
(t4) = 0.51 + 0.16i + 0.23j + 0.10k
7.16
5.03
Predicted results of traditional Markov chain-SPA
(t4) = 0.44 + 0.15i + 0.19j + 0.06k + 0.16l
6.29
16.62
Patient i
Actual results
(t4) = 0.51 + 0.30i + 0.08j + 0.11k
7.42
Predicted results of higher-order Markov chain-SPA
(t4) = 0.45 + 0.22i + 0.24j + 0.10k
7.03
5.27
Predicted results of traditional Markov chain-SPA
(t4) = 0.46 + 0.19i + 0.14j + 0.05k + 0.15l
6.53
12.01
Patient j
Actual results
(t4) = 0.51 + 0.30i + 0.19j
7.64
Predicted results of higher-order Markov chain-SPA
(t4) = 0.42 + 0.34i + 0.16j + 0.08k
7.20
5.72
Predicted results of traditional Markov chain-SPA
(t4) = 0.37 + 0.35i + 0.08j + 0.02k + 0.18l
6.39
16.35
Patient k
Actual results
(t4) = 0.64 + 0.17i + 0.19j
7.90
Predicted results of higher-order Markov chain-SPA
(t4) = 0.43 + 0.40i + 0.15j + 0.03k
7.46
5.61
Predicted results of traditional Markov chain-SPA
(t4) = 0.44 + 0.32i + 0.08j + 0.01k + 0.15l
6.80
13.89
Patient l
Actual results
(t4) = 0.39 + 0.34i + 0.27j
7.24
Predicted results of higher-order Markov chain-SPA
(t4) = 0.40 + 0.26i + 0.25j + 0.09k
6.91
4.53
Predicted results of traditional Markov chain-SPA
(t4) = 0.37 + 0.28i + 0.16j + 0.05k + 0.14l
6.39
11.78
Patient m
Actual results
(t4) = 0.21 + 0.68i + 0.11k
6.98
Predicted results of higher-order Markov chain-SPA
(t4) = 0.24 + 0.39i + 0.30j + 0.07k
6.60
5.44
Predicted results of traditional Markov chain-SPA
(t4) = 0.24 + 0.25i + 0.27j + 0.02k + 0.21l
5.60
19.79
Patient n
Actual results
(t4) = 0.37 + 0.29i + 0.34j
7.06
Predicted results of higher-order Markov chain-SPA
(t4) = 0.29 + 0.35i + 0.29j + 0.07k
6.73
4.71
Predicted results of traditional Markov chain-SPA
(t4) = 0.34 + 0.31i + 0.23j + 0.04k + 0.07l
6.60
6.53
Patient o
Actual results
(t4) = 0.38 + 0.39i + 0.23j
7.30
Predicted results of higher-order Markov chain-SPA
(t4) = 0.33 + 0.34i + 0.23j + 0.10k
6.83
6.51
Predicted results of traditional Markov chain-SPA
(t4) = 0.34 + 0.33i + 0.18j + 0.05k + 0.10l
6.51
10.80
The connection degrees predicted by higher-order Markov chain-SPA model and efficacy scores at time t4 were compared with the actual connection degrees and the connection degrees predicted by the traditional first-order Markov chain-SPA model. The relative error is the proportion of the difference between the predicted value and the actual value in the predicted value.
