文献速递头颈技术与安全CT

基于双能CT的深度学习模型评估口腔鳞状细胞癌颈部淋巴结转移

Deep Learning Model Based on Dual-energy CT for Assessing Cervical Lymph Node Metastasis in Oral Squamous Cell Carcinoma.
PubMed 文献 · Acad Radiol · 2025-07-07

本研究利用双能CT四种序列图像构建融合四通道复合图像,开发16种深度学习模型评估口腔鳞状细胞癌颈部淋巴结转移,其中Crossformer_Transformer模型诊断性能最佳,在两个验证集上AUC均为0.881,均高于放射科医师的平均AUC(0.723-0.819)。

要点

  • Crossformer_Transformer模型在外部验证集预测颈部淋巴结转移的AUC为0.881(95% CI 0.8033-0.9590)。
  • 放射科医师在两个验证集的平均AUC为0.723-0.819,低于模型。
  • 模型基于双能CT碘图、脂肪图、70 keV单能图和RHO/Z图融合图像构建。

核心结论

  • Crossformer_Transformer模型在多中心数据上验证,在口腔鳞状细胞癌患者颈部淋巴结转移的术前风险评估和临床决策方面显示出较强潜力。

研究设计与样本

  • 研究收集两个中心的术前双能CT图像(碘图、脂肪图、70 keV单能图、RHO/Z图)及临床数据;第一中心248例患者按8:2分为训练集(n=198)和内部验证集(n=50),第二中心106例患者为外部验证集;构建16种深度学习模型并与放射科医师评估比较。

诊断标准与关键数值

  • Crossformer_Transformer模型预测口腔鳞状细胞癌颈部淋巴结转移的AUC在训练集为0.960(95% CI 0.9355-0.9842),内部验证集为0.881(95% CI 0.7396-1.0000),外部验证集为0.881(95% CI 0.8033-0.9590)。
  • 放射科医师在两个验证集的平均AUC为0.723-0.819,低于模型。

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

BACKGROUND: Accurate detection of lymph node metastasis (LNM) in oral squamous cell carcinoma (OSCC) is crucial for treatment planning. This study developed a deep learning model using dual-energy CT to improve LNM detection.

METHODS: Preoperative dual-energy CT images (Iodine Map, Fat Map, monoenergetic 70 keV, and RHO/Z Map) and clinical data were collected from two centers. From the first center, 248 patients were divided into training (n=198) and internal validation (n=50) cohorts (8:2 ratio), while 106 patients from the second center comprised the external validation cohort. Region-of-interest images from all four sequences were stacked along the channel dimension to generate fused four-channel composite images. 16 deep learning models were developed as follows: three architectures (Crossformer, Densenet169, Squeezenet1_0) applied to each single-sequence/fused image, followed by MLP integration. Additionally, a Crossformer_Transformer model was constructed based on fused image. The top-performing model was compared against radiologists' assessments.

RESULTS: Among the 16 deep learning models trained in this study, the Crossformer_Transformer model demonstrated the best diagnostic performance in predicting LNM in OSCC patients, with an AUC of 0.960 (95% CI: 0.9355-0.9842) on the training dataset, and 0.881 (95% CI: 0.7396-1.0000) and 0.881 (95% CI: 0.8033-0.9590) on the internal and external validation sets, respectively. The average AUC for radiologists across both validation cohorts (0.723-0.819) was lower than that of the model.

CONCLUSION: The Crossformer_Transformer model, validated on multicenter data, shows strong potential for improving preoperative risk assessment and clinical decision-making in cervical LNM for OSCC patients.

来源:Acad Radiol,查看原文。版权归原作者所有。

引用

  1. Qi YM, Zhang LJ, Wang Y, et al. Deep Learning Model Based on Dual-energy CT for Assessing Cervical Lymph Node Metastasis in Oral Squamous Cell Carcinoma. Acad Radiol. 2025. PMID: 40628643. doi:10.1016/j.acra.2025.06.020

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。