Table 1 Accuracy (micro-averaged F1 score), macro-averaged F1 score, and weighted-averaged F1 score of mosquito classification in species, sex, and abdomen status using 5-fold cross validation.

From: Towards transforming malaria vector surveillance using VectorBrain: a novel convolutional neural network for mosquito species, sex, and abdomen status identifications

Results

Accuracy (avg (min–max))

Macro-averaged F1 score (avg (min–max))

Weighted-averaged F1 score (avg (min–max))

Species

94.44 (93.86–96.04)%

94.10 (92.43–95.20)%

94.40 (92.12–96.00)%

Sex

97.66 (96.21–98.09)%

96.17 (94.77–96.950)%

97.67 (96.20–98.30)%

Abdomen

82.20 (79.80–84.55)%

81.17 (77.03–83.56)%

81.98 (79.54–83.98)%

Back to article page