Development and validation of an endoscopic images-based deep learning model for detection with nasopharyngeal malignancies.
Development and validation of an endoscopic images-based deep learning model for detection with nasopharyngeal malignancies.
复制标题
基于内窥镜图像的深度学习模型的开发和验证，用于检测鼻咽恶性肿瘤
DOI：
10.1186/s40880-018-0325-9
复制
发表时间：
2018-09-25
期刊：
影响因子：
--
通讯作者：
Lv X
中科院分区：
文献类型：
--
作者：
Li C;Jing B;Ke L;Li B;Xia W;He C;Qian C;Zhao C;Mai H;Chen M;Cao K;Mo H;Guo L;Chen Q;Tang L;Qiu W;Yu Y;Liang H;Huang X;Liu G;Li W;Wang L;Sun R;Zou X;Guo S;Huang P;Luo D;Qiu F;Wu Y;Hua Y;Liu K;Lv S;Miao J;Xiang Y;Sun Y;Guo X;Lv X
背景由于鼻咽部解剖位置隐蔽，腺样体增生频繁，活检时恶性肿瘤诊断的阳性率较低，从而导致鼻咽部恶性肿瘤的早期诊断延迟或漏诊。本文旨在开发一种基于深度学习的鼻咽癌内窥镜检查人工智能检测工具。方法建立了一种基于内窥镜图像的鼻咽部恶性肿瘤检测模型(eNPM-DM)，该模型由基于初始结构的完全卷积网络组成，并使用单独的训练集和验证集进行分类和分割。简单地说，总共收集了28,966张合格的图像。在这些图像中，从2008年1月1日到2016年12月31日期间从7951人获得的27536张经活检证实的图像，通过简单的随机化被分成训练、验证和测试集，比例为7：1：2。此外，从2017年1月1日至2017年3月31日获得的1430张图像被用作前瞻性测试集，以比较所建立的模型的性能与肿瘤学家的评估。结果所有图像均经组织病理证实，其中正常对照5713例(19.7%)，鼻咽癌19 107例(66.0%)，鼻咽癌335例(1.2%)，良性病变3811例(13.2%)。在检测测试集中的恶性肿瘤时，eNPM-DM的总体准确率为88.7%(95%可信区间为87.8%-89.5%)。在前瞻性比较阶段，eNPM-DM优于专家：总体准确率为88.0%(95%可信区间86.1%~89.6%)vs.80.5%(95%可信区间77.0%~84.0%)。ENPM-DM算法所需时间较短(40 S vs.110.0±5.8min)，在自动分割鼻咽恶性区域方面表现出良好的性能，测试集和预期测试集的平均分割结果分别为0.78±0.24和0.75±0.26。结论eNPM-DM在鼻咽部肿块良恶性的诊断分类中优于肿瘤学家的评价，实现了从鼻咽内窥镜图像背景中自动分割恶性区域。
BackgroundDue to the occult anatomic location of the nasopharynx and frequent presence of adenoid hyperplasia, the positive rate for malignancy identification during biopsy is low, thus leading to delayed or missed diagnosis for nasopharyngeal malignancies upon initial attempt. Here, we aimed to develop an artificial intelligence tool to detect nasopharyngeal malignancies under endoscopic examination based on deep learning.MethodsAn endoscopic images‐based nasopharyngeal malignancy detection model (eNPM‐DM) consisting of a fully convolutional network based on the inception architecture was developed and fine‐tuned using separate training and validation sets for both classification and segmentation. Briefly, a total of 28,966 qualified images were collected. Among these images, 27,536 biopsy‐proven images from 7951 individuals obtained from January 1st, 2008, to December 31st, 2016, were split into the training, validation and test sets at a ratio of 7:1:2 using simple randomization. Additionally, 1430 images obtained from January 1st, 2017, to March 31st, 2017, were used as a prospective test set to compare the performance of the established model against oncologist evaluation. The dice similarity coefficient (DSC) was used to evaluate the efficiency of eNPM‐DM in automatic segmentation of malignant area from the background of nasopharyngeal endoscopic images, by comparing automatic segmentation with manual segmentation performed by the experts.ResultsAll images were histopathologically confirmed, and included 5713 (19.7%) normal control, 19,107 (66.0%) nasopharyngeal carcinoma (NPC), 335 (1.2%) NPC and 3811 (13.2%) benign diseases. The eNPM‐DM attained an overall accuracy of 88.7% (95% confidence interval (CI) 87.8%–89.5%) in detecting malignancies in the test set. In the prospective comparison phase, eNPM‐DM outperformed the experts: the overall accuracy was 88.0% (95% CI 86.1%–89.6%) vs. 80.5% (95% CI 77.0%–84.0%). The eNPM‐DM required less time (40 s vs. 110.0 ± 5.8 min) and exhibited encouraging performance in automatic segmentation of nasopharyngeal malignant area from the background, with an average DSC of 0.78 ± 0.24 and 0.75 ± 0.26 in the test and prospective test sets, respectively.ConclusionsThe eNPM‐DM outperformed oncologist evaluation in diagnostic classification of nasopharyngeal mass into benign versus malignant, and realized automatic segmentation of malignant area from the background of nasopharyngeal endoscopic images.
登录
查看更多内容
复发性腺样体切除术后小儿鼻咽癌的诊断。
影响因子：
0.6
作者：
Cengiz K;Kumral TL;Yıldırım G
通讯作者：
Yıldırım G
影响因子：
4.8
作者：
Schirrmeister, Robin Tibor;Springenberg, Jost Tobias;Ball, Tonio
通讯作者：
Ball, Tonio
影响因子：
19.7
作者：
Lakhani, Paras;Sundaram, Baskaran
通讯作者：
Sundaram, Baskaran
影响因子：
19.5
作者：
Russakovsky, Olga;Deng, Jia;Fei-Fei, Li
通讯作者：
Fei-Fei, Li
影响因子：
64.8
作者：
Esteva A;Kuprel B;Novoa RA;Ko J;Swetter SM;Blau HM;Thrun S
通讯作者：
Thrun S