文献速递头颈技术与安全US

一种改进的用于颈部淋巴结病变表征的诊断性深度学习模型

An Improved Diagnostic Deep Learning Model for Cervical Lymphadenopathy Characterization.
PubMed 文献 · Ultrasound Med Biol · 2025-07-21

本研究回顾性纳入210例有明确病理结果的颈部淋巴结病变,按4:1随机分为训练组169例和测试组41例,开发了基于B型超声的深度学习模型,并与两种传统超声诊断模型比较。测试组中深度学习模型AUC为0.871,显著高于模型I的0.681和模型II的0.679;模型III特异度为93.3%,显著高于模型I的40.0%和模型II的60.0%。结论认为基于B型超声的深度学习可能是鉴别良恶性颈部淋巴结病变的稳健工具。

要点

  • 深度学习阳性结果是恶性颈部淋巴结病变最重要的相关因素(OR=39.05)。
  • 测试组深度学习模型AUC为0.871,高于模型I的0.681和模型II的0.679。
  • 模型III特异度93.3%,高于模型I的40.0%和模型II的60.0%。

核心结论

  • 基于B型超声的深度学习可能是鉴别良恶性颈部淋巴结病变的稳健工具。

研究设计与样本

  • 回顾性纳入210例有明确病理结果的颈部淋巴结病变,按4:1随机分为训练组169例和测试组41例。
  • 开发了整合卷积神经网络、可变形卷积网络和注意力机制的深度学习模型。
  • 比较三种诊断模型:模型I为至少具有一项可疑B型超声特征即判为恶性,模型II为多因素logistic回归获得B型超声特征总风险评分,模型III为深度学习输出阳性即判为恶性。
  • 以病理结果为参考标准,采用受试者工作特征曲线下面积(AUC)及相应敏感度和特异度评估诊断效能。

关键影像表现/征象

  • 可疑B型超声特征包括纵径与短径比值<2、边缘不规则、高回声、淋巴门消失、囊性坏死和钙化。

诊断标准与关键数值

  • 训练组多因素分析显示深度学习阳性结果与恶性颈部淋巴结病变最相关(OR=39.05,p<0.001),其次为淋巴门消失(OR=6.01,p=0.001)。
  • 测试组深度学习模型AUC为0.871,显著高于模型I的0.681(p=0.04)和模型II的0.679(p=0.03)。
  • 模型III特异度为93.3%,显著高于模型I的40.0%(p=0.002)和模型II的60.0%(p=0.03)。
  • 模型I敏感度最高,但与模型III相比无显著差异(96.2% vs. 80.8%,p=0.083)。

临床应用与阅片要点

  • 深度学习模型输出阳性可作为提示恶性颈部淋巴结病变的影像学依据。
  • 淋巴门消失是与恶性颈部淋巴结病变相关的重要B型超声特征(OR=6.01)。

局限性

  • 原文未报告局限性。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

英文原文(摘要)· 点击展开

OBJECTIVE: To validate the diagnostic performance of a B-mode ultrasound-based deep learning (DL) model in distinguishing benign and malignant cervical lymphadenopathy (CLP).

METHODS: A total of 210 CLPs with conclusive pathological results were retrospectively included and separated as training (n = 169) or test cohort (n = 41) randomly at a ratio of 4:1. A DL model integrating convolutional neural network, deformable convolution network and attention mechanism was developed. Three diagnostic models were developed: (a) Model I, CLPs with at least one suspicious B-mode ultrasound feature (ratio of longitudinal to short diameter < 2, irregular margin, hyper-echogenicity, hilus absence, cystic necrosis and calcification) were deemed malignant; (b) Model II: total risk score of B-mode ultrasound features obtained by multivariate logistic regression and (c) Model III: CLPs with positive DL output are deemed malignant. The diagnostic utility of these models was assessed by the area under the receiver operating curve (AUC) and corresponding sensitivity and specificity.

RESULTS: Multivariate analysis indicated that DL positive result was the most important factor associated with malignant CLPs [odds ratio (OR) = 39.05, p < 0.001], only followed by hilus absence (OR = 6.01, p = 0.001) in the training cohort. In the test cohort, the AUC of the DL model (0.871) was significantly higher than that in model I (AUC = 0.681, p = 0.04) and model II (AUC = 0.679, p = 0.03), respectively. In addition, model III obtained 93.3% specificity, which was significantly higher than that in model I (40.0%, p = 0.002) and model II (60.0%, p = 0.03), respectively. Although the sensitivity of model I was the highest, it did not show a significant difference compared to that of model III (96.2% vs.80.8%, p = 0.083).

CONCLUSIONS: B-mode ultrasound-based DL is a potentially robust tool for the differential diagnosis of benign and malignant CLPs.

来源:Ultrasound Med Biol,查看原文。版权归原作者所有。

引用

  1. Gong W, Li M, Wang S, et al. An Improved Diagnostic Deep Learning Model for Cervical Lymphadenopathy Characterization. Ultrasound Med Biol. 2025. PMID: 40695688. doi:10.1016/j.ultrasmedbio.2025.06.025

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。