当前位置: X-MOL 学术Artif. Intell. Med. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Uncertainty-aware temporal self-learning (UATS): Semi-supervised learning for segmentation of prostate zones and beyond
Artificial Intelligence in Medicine ( IF 7.5 ) Pub Date : 2021-04-10 , DOI: 10.1016/j.artmed.2021.102073
Anneke Meyer 1 , Suhita Ghosh 1 , Daniel Schindele 2 , Martin Schostak 2 , Sebastian Stober 1 , Christian Hansen 1 , Marko Rak 1
Affiliation  

Various convolutional neural network (CNN) based concepts have been introduced for the prostate's automatic segmentation and its coarse subdivision into transition zone (TZ) and peripheral zone (PZ). However, when targeting a fine-grained segmentation of TZ, PZ, distal prostatic urethra (DPU) and the anterior fibromuscular stroma (AFS), the task becomes more challenging and has not yet been solved at the level of human performance. One reason might be the insufficient amount of labeled data for supervised training. Therefore, we propose to apply a semi-supervised learning (SSL) technique named uncertainty-aware temporal self-learning (UATS) to overcome the expensive and time-consuming manual ground truth labeling. We combine the SSL techniques temporal ensembling and uncertainty-guided self-learning to benefit from unlabeled images, which are often readily available. Our method significantly outperforms the supervised baseline and obtained a Dice coefficient (DC) of up to 78.9%, 87.3%, 75.3%, 50.6% for TZ, PZ, DPU and AFS, respectively. The obtained results are in the range of human inter-rater performance for all structures. Moreover, we investigate the method's robustness against noise and demonstrate the generalization capability for varying ratios of labeled data and on other challenging tasks, namely the hippocampus and skin lesion segmentation. UATS achieved superiority segmentation quality compared to the supervised baseline, particularly for minimal amounts of labeled data.



中文翻译:

不确定性感知时间自学习 (UATS):用于分割前列腺区域及其他区域的半监督学习

各种基于卷积神经网络 (CNN) 的概念已被引入用于前列腺的自动分割及其粗细分为过渡区 (TZ) 和外围区 (PZ)。然而,当针对 TZ、PZ、远端前列腺尿道 (DPU) 和前纤维肌肉基质 (AFS) 的细粒度分割时,该任务变得更具挑战性,并且尚未在人类表现水平上得到解决。原因之一可能是用于监督训练的标记数据量不足。因此,我们建议应用一种名为不确定性感知时间自学习 (UATS) 的半监督学习 (SSL) 技术来克服昂贵且耗时的手动地面实况标记。我们将 SSL 技术时间集成和不确定性引导的自学习相结合,以从未标记的图像中受益,通常很容易获得。我们的方法显着优于监督基线,并分别为 TZ、PZ、DPU 和 AFS 获得了高达 78.9%、87.3%、75.3%、50.6% 的 Dice 系数 (DC)。获得的结果在所有结构的人类评价者间表现范围内。此外,我们研究了该方法对噪声的鲁棒性,并展示了对不同比例的标记数据和其他具有挑战性的任务(即海马体和皮肤病变分割)的泛化能力。与监督基线相比,UATS 实现了优越的分割质量,特别是对于最少量的标记数据。分别。获得的结果在所有结构的人类评价者间表现范围内。此外,我们研究了该方法对噪声的鲁棒性,并展示了对不同比例的标记数据和其他具有挑战性的任务(即海马体和皮肤病变分割)的泛化能力。与监督基线相比,UATS 实现了优越的分割质量,特别是对于最少量的标记数据。分别。获得的结果在所有结构的人类评价者间表现范围内。此外,我们研究了该方法对噪声的鲁棒性,并展示了对不同比例的标记数据和其他具有挑战性的任务(即海马体和皮肤病变分割)的泛化能力。与监督基线相比,UATS 实现了优越的分割质量,特别是对于最少量的标记数据。

更新日期:2021-04-24
down
wechat
bug