Table 2 The segmentation performance of each model on static ultrasound images
From: A deep learning based ultrasound diagnostic tool driven by 3D visualization of thyroid nodules
DeepLabv3+ | Unet | Unet++ | Attention Unet | TNVis | |
|---|---|---|---|---|---|
DSC | 0.73 [0.57, 0.89] | 0.80 [0.51, 1.00] | 0.88 [0.74, 1.00] | 0.88 [0.74, 1.00] | 0.90 [0.76, 1.00] |
HD95 | 38.6 [12.1, 65.1] | 19.6 [0, 46.8] | 20.1 [0, 45.2] | 19.5 [0, 44.0] | 15.1 [0, 38.6] |
Acc | 0.61 [0.42, 0.80] | 0.83 [0.55, 1.00] | 0.88 [0.73, 1.00] | 0.89 [0.74, 1.00] | 0.90 [0.75, 1.00] |
Rec | 0.98 [0.90, 1.00] | 0.79 [0.49, 1.00] | 0.88 [0.72, 1.00] | 0.89 [0.74, 1.00] | 0.91 [0.76, 1.00] |
Spec | 0.80 [0.61, 0.99] | 0.89 [0.61, 1.00] | 0.96 [0.86, 1.00] | 0.97 [0.87, 1.00] | 0.97 [0.86, 1.00] |
IoU | 0.60 [0.41, 0.79] | 0.73 [0.45, 1.00] | 0.80 [0.65, 0.95] | 0.81 [0.66, 0.96] | 0.84 [0.69, 0.99] |
