Automated cervical cell segmentation using deep ensemble learning

BMC Medical Imaging

Table 5 Performance comparison of baseline and ensemble models on the testing dataset

Task Type	Model	Dice	Sensitivity	Specificity
Cytoplasm	Best Baseline Model	0.948	0.954	0.9823
Cytoplasm	Ensemble model	0.9535 (0.9534–0.9536)	0.9621 (0.9619–0.9622)	0.9835 (0.9834–0.9836)
Nucleus	Best Baseline Model	0.750	0.713	0.9988
Nucleus	Ensemble model	0.7863 (0.7851–0.7876)	0.9581 (0.9573–0.959)	0.9961 (0.9961–0.9962)

In the first column, cytoplasm and nucleus stand for the cytoplasm segmentation task and the nucleus segmentation task, respectively. Bold values represent the best results, and confidence intervals are depicted in brackets

ISSN: 1471-2342