Table 3 The Comparison of total predication accuracy of the BP-GA.

From: Warfarin maintenance dose Prediction for Patients undergoing heart valve replacement— a hybrid model with genetic algorithm and Back-Propagation neural network

Model

MAE*

RMSE*

Predicted percentage

Underestimate (%)

Ideal (%)**

Overestimate (%)

BP-GAiv

0.383 (0.365~0.401)

0.664 (0.631~0.696)

746 (21.0)

2088 (58.7)

724 (20.3)

BP-GAv

0.370 (0.342~0.397)

0.656 (0.606~0.704)

257 (17.6)

920 (62.9)

286 (19.5)

  1. Note: BP-GAiv: the internal validation group.
  2. BP-GAev: the external validation group.
  3. *P < 0.05 (independent sample t-test of the MAE and RMSE between the internal and external validation group, respectively).
  4. **P < 0.05 (chi-square test the ideal percentage between the internal and external validation group).
  5. Ideal: the percentage of patients whose predicted absolute error between predicted dose and actual dose was within 20% of actual dose.
  6. Underestimate: the percentage of patients whose predicted dose was less than actual dose and the predicted absolute error between predicted dose and actual dose was more than 20% of the actual dose.
  7. Overestimate: the percentage of patients whose predicted dose was more than actual dose and the predicted absolute error between predicted dose and actual dose was more than 20% of the actual dose.
Back to article page