IEEE Trans Biomed Eng. 2021 May;68(5):1518-1526. doi: 10.1109/TBME.2020.3042640. Epub 2021 Apr 21.
Three-dimensional (3D) blood vessel structure information is important for diagnosis and treatment in various clinical scenarios. We present a fully automatic method for the extraction and differentiation of the arterial and venous vessel trees from abdominal contrast enhanced computed tomography (CE-CT) volumes using convolutional neural networks (CNNs).
We used a novel ratio-based sampling method to train 2D and 3D versions of the U-Net, the V-Net and the DeepVesselNet. Networks were trained with a combination of the Dice and cross entropy loss. Performance was evaluated on 20 IRCAD subjects. Best performing networks were combined into an ensemble. We investigated seven different weighting schemes. Trained networks were additionally applied to 26 BTCV cases to validate the generalizability.
Based on our experiments, the optimal configuration is an equally weighted ensemble of 2D and 3D U- and V-Nets. Our method achieved Dice similarity coefficients of 0.758 ± 0.050 (veins) and 0.838 ± 0.074 (arteries) on the IRCAD data set. Application to the BTCV data set showed a high transfer ability.
Abdominal vascular structures can be segmented more accurately using ensembles than individual CNNs. 2D and 3D networks have complementary strengths and weaknesses. Our ensemble of 2D and 3D U-Nets and V-Nets in combination with ratio-based sampling achieves a high agreement with manual annotations for both artery and vein segmentation. Our results surpass other state-of-the-art methods.
Our segmentation pipeline can provide valuable information for the planning of living donor organ transplantations.
三维（3D）血管结构信息对于各种临床情况下的诊断和治疗都很重要。我们提出了一种从腹部增强计算机断层扫描（CE-CT）容积中提取和区分动静脉树的全自动方法，该方法使用卷积神经网络（CNN）。
我们使用了一种新的基于比率的采样方法来训练二维和三维版本的 U-Net、V-Net 和 DeepVesselNet。网络使用 Dice 和交叉熵损失的组合进行训练。在 20 个 IRCAD 受试者上评估了性能。表现最好的网络被组合成一个集合。我们研究了七种不同的加权方案。训练好的网络还应用于 26 个 BTCV 案例，以验证其通用性。
根据我们的实验，最佳配置是二维和三维 U-和 V-Net 的等权重集合。我们的方法在 IRCAD 数据集上实现了 0.758±0.050（静脉）和 0.838±0.074（动脉）的 Dice 相似系数。应用于 BTCV 数据集显示出较高的转移能力。
与单个 CNN 相比，使用集合可以更准确地分割腹部血管结构。二维和三维网络具有互补的优缺点。我们的二维和三维 U-Net 和 V-Net 的集合与基于比率的采样相结合，在动脉和静脉分割方面与手动注释具有高度一致性。我们的结果超过了其他最先进的方法。
我们的分割流水线可以为活体供体器官移植的规划提供有价值的信息。
