The State Key Laboratory Breeding Base of Basic Science of Stomatology (Hubei-MOST) & Key Laboratory of Oral Biomedicine Ministry of Education, School & Hospital of Stomatology, Wuhan University, Wuhan, China.
School of Computer Science, Wuhan University, 299 Bayi Road, Wuhan, 430072, Hubei, China.
Eur Radiol. 2023 Jun;33(6):4303-4312. doi: 10.1007/s00330-022-09355-5. Epub 2022 Dec 28.
Lymph node (LN) metastasis is a common cause of recurrence in oral cancer; however, the accuracy of distinguishing positive and negative LNs is not ideal. Here, we aimed to develop a deep learning model that can identify, locate, and distinguish LNs in contrast-enhanced CT (CECT) images with a higher accuracy.
The preoperative CECT images and corresponding postoperative pathological diagnoses of 1466 patients with oral cancer from our hospital were retrospectively collected. In stage I, full-layer images (five common anatomical structures) were labeled; in stage II, negative and positive LNs were separately labeled. The stage I model was innovatively employed for stage II training to improve accuracy with the idea of transfer learning (TL). The Mask R-CNN instance segmentation framework was selected for model construction and training. The accuracy of the model was compared with that of human observers.
A total of 5412 images and 5601 images were labeled in stage I and II, respectively. The stage I model achieved an excellent segmentation effect in the test set (AP-0.7249). The positive LN accuracy of the stage II TL model was similar to that of the radiologist and much higher than that of the surgeons and students (0.7042 vs. 0.7647 (p = 0.243), 0.4216 (p < 0.001), and 0.3629 (p < 0.001)). The clinical accuracy of the model was highest (0.8509 vs. 0.8000, 0.5500, 0.4500, and 0.6658 of the Radiology Department).
The model was constructed using a deep neural network and had high accuracy in LN localization and metastasis discrimination, which could contribute to accurate diagnosis and customized treatment planning.
• Lymph node metastasis is not well recognized with modern medical imaging tools. • Transfer learning can improve the accuracy of deep learning model prediction. • Deep learning can aid the accurate identification of lymph node metastasis.
淋巴结（LN）转移是口腔癌复发的常见原因，但区分阳性和阴性 LNs 的准确性并不理想。在这里，我们旨在开发一种深度学习模型，该模型可以在对比度增强 CT（CECT）图像中以更高的准确性识别、定位和区分 LNs。
回顾性收集了我院 1466 例口腔癌患者的术前 CECT 图像及相应的术后病理诊断。在第一阶段，对全层图像（五个常见解剖结构）进行标记；在第二阶段，分别标记阴性和阳性 LNs。创新性地采用第一阶段的模型对第二阶段进行训练，以通过转移学习（TL）的思想提高准确性。选择 Mask R-CNN 实例分割框架进行模型构建和训练。比较了模型的准确性与人类观察者的准确性。
第一阶段和第二阶段分别标记了 5412 张和 5601 张图像。第一阶段模型在测试集中的分割效果较好（AP-0.7249）。第二阶段 TL 模型的阳性 LN 准确率与放射科医生相似，且明显高于外科医生和学生（0.7042 与 0.7647（p=0.243），0.4216（p<0.001），0.3629（p<0.001））。模型的临床准确率最高（放射科为 0.8509，分别为 0.8000、0.5500、0.4500 和 0.6658）。
该模型使用深度神经网络构建，具有较高的 LN 定位和转移判别准确性，有助于准确诊断和制定个体化治疗计划。
现代医学成像工具对淋巴结转移的识别效果不佳。
转移学习可以提高深度学习模型预测的准确性。
深度学习可辅助准确识别淋巴结转移。
Zotero 插件
