An Edge-Assisted Computing and Mask Attention Based Network for Lung Region Segmentation

Table 6

Results of EAM-Net and the state-of-the-art lung region segmentation methods on the Montgomery dataset.

Methods	PA (%)	Dice (%)	JA (%)
Feature selection with BN [10]	78.06	62.31	43.97
Feature selection with MLP [10]	79.16	64.17	46.04
Feature selection with RF [10]	80.81	66.32	49.27
Feature selection and vote [10]	83.44	69.89	53.72
Bayesian feature pyramid network [21]	96.19	93.07	87.04
Souza et al. [8]	97.01	94.12	88.27
Rahman et al. [18]	96.84	94.25	89.13
ET-Net [20]	98.51	97.29	94.32
CFCM18 [24]	98.19	96.67	93.61
CFCM34 [24]	98.30	96.91	93.99
CFCM50 [24]	98.35	97.01	94.17
CFCM101 [24]	98.46	97.18	94.55
Yahyatabar et al. [17]	98.52	97.30	94.74
X-ray-Net [23]	98.57	97.44	94.93
EAM-Net	99.16	98.23	96.52

Bold values represent the the highest performance for each performance metric.