甲状腺癌中存在明显的甲状腺外扩展（ETE）会影响患者的预后，但成像检查无法可靠诊断。本研究旨在开发深度学习（DL）模型，以在手术前通过超声图像定位和评估甲状腺癌结节是否存在明显的ETE。回顾性分析了4个医疗中心2016年1月至2021年12月的806个甲状腺癌结节（4451张图像）的灰度超声图像，其中包括517个未发生明显ETE结节和289个发生明显ETE结节。从内部数据集中随机选择了283个未发生明显ETE结节和158个发生明显ETE结节组成训练集和验证集（2914张图像），并构建了一个多任务DL模型以诊断明显ETE。此外，还建立了临床模型和临床与DL组合模型。在内部测试集（974张图像，其中包括139个未发生明显ETE结节和83个发生明显ETE结节）和外部测试集（563张图像，其中包括95个未发生明显ETE结节和48个发生明显ETE结节）中，根据病理结果验证了DL模型的诊断性能。然后将结果与由2名高级和2名初级放射科医师诊断的结果进行比较。在内部测试集中，DL模型表现出最高的AUC（0.91；95% CI：0.87、0.96），显著高于两名高级放射科医师的AUC（0.78；95% CI：0.71、0.85；P < 0.001）和（0.76；95% CI：0.70、0.83；P < 0.001）以及两名初级放射科医师的AUC（0.65；95% CI：0.58、0.73；P < 0.001）和（0.69；95% CI：0.62、0.77；P < 0.001）。DL模型显著高于临床模型（0.84；95% CI：0.79、0.89；P = 0.019），但DL模型和临床及DL组合模型之间没有显著差异（AUC：0.94；95% CI：0.91、0.97；P = 0.143）。在外部测试集中，DL模型也表现出最高的AUC（0.88；95% CI：0.81、0.94），显著高于一名高级放射科医师的AUC（0.75；95% CI：0.66、0.84；P = 0.008）和（0.81；95% CI：0.72、0.89；P = 0.152）以及两名初级放射科医师的AUC（0.72；95% CI：0.62、0.81；P = 0.002）和（0.67；95% CI：0.57、0.77；P < 0.001）。DL模型和临床模型之间没有显著差异（AUC：0.85；95% CI：0.79、0.91；P = 0.516）和临床+DL模型（AUC：0.92；95% CI：0.87、0.96；P = 0.093）。使用DL模型，两名初级放射科医师的诊断能力显著提高。基于超声成像的DL模型是一个简单而有用的工具，用于甲状腺癌的明显ETE的术前诊断，其诊断性能等同于或甚至优于高级放射科医师。江西省自然科学基金（20224BAB216079），江西省重点研发计划（20181BBG70031）和南昌大学自然科学交叉创新基金（9167-28220007-YB2110）。© 2023 The Author(s)。

The presence of gross extrathyroidal extension (ETE) in thyroid cancer will affect the prognosis of patients, but imaging examination cannot provide a reliable diagnosis for it. This study was conducted to develop a deep learning (DL) model for localization and evaluation of thyroid cancer nodules in ultrasound images before surgery for the presence of gross ETE.From January 2016 to December 2021 grayscale ultrasound images of 806 thyroid cancer nodules (4451 images) from 4 medical centers were retrospectively analyzed, including 517 no gross ETE nodules and 289 gross ETE nodules. 283 no gross ETE nodules and 158 gross ETE nodules were randomly selected from the internal dataset to form a training set and validation set (2914 images), and a multitask DL model was constructed for diagnosing gross ETE. In addition, the clinical model and the clinical and DL combined model were constructed. In the internal test set [974 images (139 no gross ETE nodules and 83 gross ETE nodules)] and the external test set [563 images (95 no gross ETE nodules and 48 gross ETE nodules)], the diagnostic performance of DL model was verified based on the pathological results. And then, compared the results with the diagnosis by 2 senior and 2 junior radiologists.In the internal test set, DL model demonstrated the highest AUC (0.91; 95% CI: 0.87, 0.96), which was significantly higher than that of two senior radiologists [(AUC, 0.78; 95% CI: 0.71, 0.85; P < 0.001) and (AUC, 0.76; 95% CI: 0.70, 0.83; P < 0.001)] and two juniors radiologists [(AUC, 0.65; 95% CI: 0.58, 0.73; P < 0.001) and (AUC, 0.69; 95% CI: 0.62, 0.77; P < 0.001)]. DL model was significantly higher than clinical model [(AUC, 0.84; 95% CI: 0.79, 0.89; P = 0.019)], but there was no significant difference between DL model and clinical and DL combined model [(AUC, 0.94; 95% CI: 0.91, 0.97; P = 0.143)]. In the external test set, DL model also demonstrated the highest AUC (0.88, 95% CI: 0.81, 0.94), which was significantly higher than that of one of senior radiologists [(AUC, 0.75; 95% CI: 0.66, 0.84; P = 0.008) and (AUC, 0.81; 95% CI: 0.72, 0.89; P = 0.152)] and two junior radiologists [(AUC, 0.72; 95% CI: 0.62, 0.81; P = 0.002) and (AUC, 0.67; 95 CI: 0.57, 0.77; P < 0.001]. There was no significant difference between DL model and clinical model [(AUC, 0.85; 95% CI: 0.79, 0.91; P = 0.516)] and clinical + DL model [(AUC, 0.92; 95% CI: 0.87, 0.96; P = 0.093)]. Using DL model, the diagnostic ability of two junior radiologists was significantly improved.The DL model based on ultrasound imaging is a simple and helpful tool for preoperative diagnosis of gross ETE thyroid cancer, and its diagnostic performance is equivalent to or even better than that of senior radiologists.Jiangxi Provincial Natural Science Foundation (20224BAB216079), the Key Research and Development Program of Jiangxi Province (20181BBG70031), and the Interdisciplinary Innovation Fund of Natural Science, Nanchang University (9167-28220007-YB2110).© 2023 The Author(s).