Table 2 Ablation experiment of MD-Net on the Vid-QU-EX dataset (Bold represents the best result).

From: Multi-scale input layers and dense decoder aggregation network for COVID-19 lesion segmentation from CT scans

Method

Dice

Mcc

Jaccard

Baseline (U-Net)

0.8265

0.7992

0.7051

Baseline+MIL

0.8333

0.8074

0.7155

Baseline+SE-Conv

0.8350

0.8090

0.7180

Baseline+DDA

0.8305

0.8046

0.7114

Baseline+MIL+DDA

0.8377

0.8127

0.7217

Baseline+MIL+SE-Conv

0.8404

0.8152

0.7257

Baseline+SE-Conv+DDA

0.8360

0.8106

0.7197

Baseline+MIL+DDA+SE-Conv (MD-Net)

0.8425

0.8176

0.7292

Back to article page