Table 7 Paired t-test results (p-values).

From: Optimizing diabetic retinopathy detection with electric fish algorithm and bilinear convolutional networks

Comparison

p-value (Accuracy)

BCAN vs. DCNN

0.003

BCAN vs. MuR-CAN

0.007

BCAN vs. MAPCRCI-DMPLC

0.001

  1. Wilcoxon signed-rank test results are as follows: BCAN vs. DCNN: p = 0.004. BCAN vs. MuR-CAN: p = 0.009.
  2. These results indicate that the performance improvement of BCAN is statistically significant (p < 0.05) compared to the baseline models.
Back to article page