Lung Infection Segmentation for COVID-19 Pneumonia Based on a Cascade Convolutional Network from CT Images

Table 2

Quantitative comparison of infected tissue segmentation outcomes based on our model and three recently published structures. The evaluations are based on average surface distance (ASD), relative volume difference (RVD), Volume overlap error (VOE), root mean square symmetric surface distance (RMS), and maximum surface distance (MSD).

Approach	ASD (mm)	VOE (%)	RVD (%)	MSD (mm)	RMS (mm)
DenseNet201 [1]
Weakly supervised deep learning [71]
Weakly supervised framework [4]
Proposed CNN
Proposed CNN+LDN
Proposed CNN+fuzzy -means
Proposed CNN+fuzzy -means+LDN