Deep Learning for Automatic Segmentation of Oral and Oropharyngeal Cancer Using Narrow Band Imaging: Preliminary Experience in a Clinical Perspective.
Deep Learning for Automatic Segmentation of Oral and Oropharyngeal Cancer Using Narrow Band Imaging: Preliminary Experience in a Clinical Perspective.
复制标题
DOI：
10.3389/fonc.2021.626602
复制
发表时间：
2021
影响因子：
4.7
通讯作者：
Moccia S
中科院分区：
文献类型：
--
作者：
Paderno A;Piazza C;Del Bon F;Lancini D;Tanagli S;Deganello A;Peretti G;De Momi E;Patrini I;Ruperti M;Mattos LS;Moccia S
应用于视频分析的全卷积神经网络（FCNN）在头颈部肿瘤学领域特别有趣，因为内窥镜检查是上消化道癌症患者诊断、分期和随访的关键步骤。本研究的目的是测试基于fcnn的口腔（OC）和口咽部（OP）鳞状细胞癌（SCC）的语义分割方法。从一家三级学术医院的机构登记处检索了两个数据集，分别分析了34个和45个NBI内窥镜下的OC和OP病变。OC的数据集由110帧组成，OP的数据集由116帧组成。研究了三种fcnn （U-Net、U-Net 3和ResNet）对肿瘤图像的分割。对每个测试网络的fcnn性能进行评估，并与金标准（由专家临床医生进行的手动注释表示）进行比较。对于基于fcnn的OC数据集分割，使用5个（×2）块和16个滤波器的ResNet方法在Dice Similarity Coefficient （Dsc）方面取得了最好的结果，中位数为0.6559。在基于fcnn的OP数据集分割中，使用4（×2）块和16个滤波器的ResNet在Dsc方面取得了最好的结果，中位数为0.7603。所有测试的fcnn呈现非常高的方差值，导致所有评估指标的极小值非常低。fcnn在OC和OP视频内窥镜图像的分析和分割方面具有广阔的应用前景。所有测试的FCNN架构在诊断准确性方面都显示出令人满意的结果。处理网络的推理时间特别短，在14 ~ 115 ms之间，显示了实时应用的可能性。
Fully convoluted neural networks (FCNN) applied to video-analysis are of particular interest in the field of head and neck oncology, given that endoscopic examination is a crucial step in diagnosis, staging, and follow-up of patients affected by upper aero-digestive tract cancers. The aim of this study was to test FCNN-based methods for semantic segmentation of squamous cell carcinoma (SCC) of the oral cavity (OC) and oropharynx (OP). Two datasets were retrieved from the institutional registry of a tertiary academic hospital analyzing 34 and 45 NBI endoscopic videos of OC and OP lesions, respectively. The dataset referring to the OC was composed of 110 frames, while 116 frames composed the OP dataset. Three FCNNs (U-Net, U-Net 3, and ResNet) were investigated to segment the neoplastic images. FCNNs performance was evaluated for each tested network and compared to the gold standard, represented by the manual annotation performed by expert clinicians. For FCNN-based segmentation of the OC dataset, the best results in terms of Dice Similarity Coefficient (Dsc) were achieved by ResNet with 5(×2) blocks and 16 filters, with a median value of 0.6559. In FCNN-based segmentation for the OP dataset, the best results in terms of Dsc were achieved by ResNet with 4(×2) blocks and 16 filters, with a median value of 0.7603. All tested FCNNs presented very high values of variance, leading to very low values of minima for all metrics evaluated. FCNNs have promising potential in the analysis and segmentation of OC and OP video-endoscopic images. All tested FCNN architectures demonstrated satisfying outcomes in terms of diagnostic accuracy. The inference time of the processing networks were particularly short, ranging between 14 and 115 ms, thus showing the possibility for real-time application.
登录
查看更多内容
影响因子：
5.3
作者：
Kono, Mitsuhiro;Ishihara, Ryu;Tada, Tomohiro
通讯作者：
Tada, Tomohiro
影响因子：
0.8
作者：
Cernazanu-Glavan, Cosmin;Holban, Stefan
通讯作者：
Holban, Stefan
DOI：
10.1002/hed.23582
发表时间：
2015-02-01
影响因子：
2.9
作者：
Bertino, Giulia;Cacciola, Salvatore;Benazzo, Marco
通讯作者：
Benazzo, Marco
DOI：
10.1002/hed.26313
发表时间：
2020-06-16
影响因子：
2.9
作者：
Inaba, Atsushi;Hori, Keisuke;Yano, Tomonori
通讯作者：
Yano, Tomonori
影响因子：
2.6
作者：
Ni, Xiao-Guang;Zhu, Ji-Qing;Wang, Gui-Qi
通讯作者：
Wang, Gui-Qi