Skip to main content

Table 3 Image-level with union vs Image-level without union

From: Image-level and group-level models for Drosophilagene expression pattern annotation

Stages Without union With union
  Accuracy AUC Sensitivity Specificity Accuracy AUC Sensitivity Specificity
4-6 69.49 0.6796 0.6623 0.6968 77.35 0.7437 0.7002 0.7872
7-8 75.78 0.7509 0.7260 0.7758 78.42 0.7666 0.7209 0.8123
9-10 74.39 0.7469 0.7373 0.7565 78.98 0.7592 0.7481 0.7702
11-12 74.69 0.7574 0.7753 0.7395 79.84 0.8021 0.8153 0.7889
13-17 78.24 0.7687 0.7433 0.7939 82.04 0.8012 0.7655 0.8359
  1. Image-level with union performs much better than without union (sparse coding).