Table 2 Quantitative results from the ablation study. The results were obtained without conducting self-training. The paired t-test was conducted to assess the statistical significance of our results. \((*p< 0.05, **p < 0.0001)\).

From: Target-aware cross-modality unsupervised domain adaptation for vestibular schwannoma and cochlea segmentation

Method

Dice score (\(\uparrow\))

ASSD (\(\downarrow\))

VS

Cochlea

Mean

VS

Cochlea

CycleGAN33

0.7798 (±0.1901)*

0.8066 (±0.0323)**

0.7932 (±0.0972)

0.8750 (±0.9222)*

0.2422 (±0.1608)**

CUT16

0.7693 (±0.2097)*

0.7950 (±0.0337)**

0.7822 (±0.1095)

0.7126 (±0.6179)*

0.2586 (±0.1668)**

QS-Attn18

0.7779 (±0.1825)*

0.8158 (±0.0287)**

0.7968 (±0.0929)

0.6667 (±0.3891)*

0.2365 (±0.1573)**

Multi-view17

0.8043 (±0.1656)

0.8158 (±0.0289)**

0.8101 (±0.0863)

0.5742 (±0.2461)*

0.2387 (±0.1581)**

Ours (w/o. RAIN)

0.8017 (±0.1696)

0.8372 (±0.0304)*

0.8195 (±0.0896)

1.3809 (±6.5339)

0.2070 (±0.1608)*

Ours (w/o. SegDecoder)

0.8057 (±0.1356)

0.8393 (±0.0299)

0.8225 (±0.0716)

1.7198 (±6.5030)*

0.2064 (±0.1597)

Ours

0.8097 (±0.1635)

0.8397 (±0.0293)

0.8247 (±0.0863)

0.5708 (±0.2919)

0.2042 (±0.1610)

Back to article page