Development of in-house fully residual deep convolutional neural network-based segmentation software for the male pelvic CT.
Development of in-house fully residual deep convolutional neural network-based segmentation software for the male pelvic CT.
复制标题
DOI：
10.1186/s13014-021-01867-6
复制
发表时间：
2021-07-22
期刊：
影响因子：
--
通讯作者：
Mizowaki T
中科院分区：
文献类型：
--
作者：
Hirashima H;Nakamura M;Baillehache P;Fujimoto Y;Nakagawa S;Saruya Y;Kabasawa T;Mizowaki T
本研究旨在（1）开发一种基于完全残差深度卷积神经网络（CNN）的分割软件，用于男性骨盆区域的计算机断层扫描图像分割;（2）证明其在男性骨盆区域的有效性。共招募了470名接受过调强放疗或调容弧形治疗的前列腺癌患者。我们的模型基于FusionNet，这是一种完全残留的深度CNN，用于对生物图像进行语义分割。为了开发基于CNN的分割软件，随机选择了450名患者，并将其分为训练组、验证组和测试组（分别为270名、90名和90名患者）。在实验1中，为了确定最佳模型，我们首先根据训练数据集的大小（90，180和270名患者）评估分割精度。在实验2中，评估了不同训练标签数量对分割精度的影响。在确定最佳模型后，在实验3中，将开发的软件用于剩余的20个数据集以评估分割精度。在实验3中，计算体积骰子相似系数（DSC）和第95百分位数Hausdorff距离（95%HD），以评估每个器官的分割准确性。在实验1中，数据集1（90例患者）的前列腺中位DSC为0.61，数据集2（180例患者）为0.86，数据集3（270例患者）为0.86。当训练病例的数量从90增加到180时，所有器官的中位DSC显着增加，但从180增加到270时没有改善。在实验2中，在训练期间施加的标签的数量对DSC的影响很小。通过270例患者和4个器官建立了最佳模型。在实验3中，前列腺的DSC和95%HD值的中位数分别为0.82和3.23 mm;精囊的DSC和95%HD值的中位数分别为0.71和3.82 mm;直肠的DSC和95%HD值的中位数分别为0.89和2.65 mm;膀胱的DSC和95%HD值的中位数分别为0.95和4.18 mm。我们已经开发了一个基于CNN的男性骨盆区域分割软件，并证明了基于CNN的分割软件是有效的男性骨盆区域。在线版本包含补充材料，可通过10.1186/s13014-021-01867-6获得。
This study aimed to (1) develop a fully residual deep convolutional neural network (CNN)-based segmentation software for computed tomography image segmentation of the male pelvic region and (2) demonstrate its efficiency in the male pelvic region. A total of 470 prostate cancer patients who had undergone intensity-modulated radiotherapy or volumetric-modulated arc therapy were enrolled. Our model was based on FusionNet, a fully residual deep CNN developed to semantically segment biological images. To develop the CNN-based segmentation software, 450 patients were randomly selected and separated into the training, validation and testing groups (270, 90, and 90 patients, respectively). In Experiment 1, to determine the optimal model, we first assessed the segmentation accuracy according to the size of the training dataset (90, 180, and 270 patients). In Experiment 2, the effect of varying the number of training labels on segmentation accuracy was evaluated. After determining the optimal model, in Experiment 3, the developed software was used on the remaining 20 datasets to assess the segmentation accuracy. The volumetric dice similarity coefficient (DSC) and the 95th-percentile Hausdorff distance (95%HD) were calculated to evaluate the segmentation accuracy for each organ in Experiment 3. In Experiment 1, the median DSC for the prostate were 0.61 for dataset 1 (90 patients), 0.86 for dataset 2 (180 patients), and 0.86 for dataset 3 (270 patients), respectively. The median DSCs for all the organs increased significantly when the number of training cases increased from 90 to 180 but did not improve upon further increase from 180 to 270. The number of labels applied during training had a little effect on the DSCs in Experiment 2. The optimal model was built by 270 patients and four organs. In Experiment 3, the median of the DSC and the 95%HD values were 0.82 and 3.23 mm for prostate; 0.71 and 3.82 mm for seminal vesicles; 0.89 and 2.65 mm for the rectum; 0.95 and 4.18 mm for the bladder, respectively. We have developed a CNN-based segmentation software for the male pelvic region and demonstrated that the CNN-based segmentation software is efficient for the male pelvic region. The online version contains supplementary material available at 10.1186/s13014-021-01867-6.
登录
查看更多内容
影响因子：
--
作者：
Vaassen F;Hazelaar C;Vaniqui A;Gooding M;van der Heyden B;Canters R;van Elmpt W
通讯作者：
van Elmpt W
影响因子：
3.8
作者：
Tong N;Gou S;Yang S;Ruan D;Sheng K
通讯作者：
Sheng K
DOI：
10.3390/diagnostics10110959
发表时间：
2020-11-17
期刊：
Diagnostics (Basel, Switzerland)
影响因子：
--
作者：
Kiljunen T;Akram S;Niemelä J;Löyttyniemi E;Seppälä J;Heikkilä J;Vuolukka K;Kääriäinen OS;Heikkilä VP;Lehtiö K;Nikkinen J;Gershkevitsh E;Borkvel A;Adamson M;Zolotuhhin D;Kolk K;Pang EPP;Tuan JKL;Master Z;Chua MLK;Joensuu T;Kononen J;Myllykangas M;Riener M;Mokka M;Keyriläinen J
通讯作者：
Keyriläinen J
HCP：用于多标签图像分类的灵活 CNN 框架
DOI：
10.1109/tpami.2015.2491929
发表时间：
2016-09-01
影响因子：
23.6
作者：
Wei, Yunchao;Xia, Wei;Yan, Shuicheng
通讯作者：
Yan, Shuicheng
影响因子：
3.5
作者：
Balagopal, Anjali;Kazemifar, Samaneh;Jiang, Steve
通讯作者：
Jiang, Steve