Automated cervical cell segmentation using deep ensemble learning

BMC Medical Imaging

Table 4 Performance comparison of ensemble models and the arithmetic means of base models on the testing dataset

Task Type	Model	Dice	Sensitivity	Specificity
Cytoplasm	Ensemble model	0.9535 (0.9534–0.9536)	0.9621 (0.9619–0.9622)	0.9835 (0.9834–0.9836)
	Average value	0.9521	0.9601	0.9830
Nucleus	Ensemble model	0.7863 (0.7851–0.7876)	0.9581 (0.9573–0.959)	0.9961 (0.9961–0.9962)
	Average value	0.7676	0.9533	0.9957

In the first column, cytoplasm and nucleus stand for the cytoplasm segmentation task and the nucleus segmentation task, respectively. For every task, the first row depicts performance metrics of the ensemble model and the second row depicts the average performance metrics of base models. Bold values represent the best results, and confidence intervals are depicted in brackets

ISSN: 1471-2342