文献速递甲状腺与浅表头颈US

人工智能在提高TI-RADS 4类结节诊断准确性中的价值

Value of Artificial Intelligence in Improving the Accuracy of Diagnosing TI-RADS Category 4 Nodules.
PubMed 文献 · Ultrasound Med Biol · 2023-08-30

本研究回顾性收集2019年1月至2022年9月1117个病理明确的TI-RADS 4类结节,并另用125个结节的外部数据集测试,比较传统机器学习、深度卷积神经网络及融合模型与不同经验放射科医师的鉴别诊断性能。外部测试集中融合模型AUROC为0.880,高于有经验放射科医师的0.815,深度卷积神经网络模型AUROC为0.837至0.856,传统机器学习模型为0.767至0.790。

要点

  • 外部测试集中融合模型诊断TR4结节的AUROC为0.880,高于有经验放射科医师的0.815。
  • 三种深度卷积神经网络模型AUROC为InceptionV3 0.852、DenseNet121 0.837、ResNet50 0.856。
  • 三种传统机器学习模型AUROC为逻辑回归0.782、多层感知机0.790、随机森林0.767。

核心结论

  • 将人工智能算法整合到医学影像研究中可能提高术前识别高风险TR4结节的准确性,并具有显著的临床应用潜力。

研究设计与样本

  • 回顾性研究,收集2019年1月至2022年9月1117个病理明确的TR4结节,另用125个TR4结节作为独立外部数据集测试;采用五折交叉验证,比较传统特征机器学习模型、深度卷积神经网络模型、融合模型及不同经验放射科医师的诊断性能。

诊断标准与关键数值

  • 外部测试集中,InceptionV3模型AUROC为0.852。
  • 外部测试集中,DenseNet121模型AUROC为0.837。
  • 外部测试集中,ResNet50模型AUROC为0.856。
  • 外部测试集中,逻辑回归模型AUROC为0.782,多层感知机模型AUROC为0.790,随机森林模型AUROC为0.767。
  • 外部测试集中,融合诊断模型AUROC为0.880,有经验放射科医师AUROC为0.815。

临床应用与阅片要点

  • 作者认为,将人工智能算法整合到医学影像研究中可能提高术前识别高风险TR4结节的准确性,并具有显著的临床应用潜力。

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: Considerable heterogeneity is observed in the malignancy rates of thyroid nodules classified as category 4 according to the Thyroid Imaging Reporting and Data System (TI-RADS). This study was aimed at comparing the diagnostic performance of artificial intelligence algorithms and radiologists with different experience levels in distinguishing benign and malignant TI-RADS 4 (TR4) nodules.

METHODS: Between January 2019 and September 2022, 1117 TR4 nodules with well-defined pathological findings were collected for this retrospective study. An independent external data set of 125 TR4 nodules was incorporated for testing purposes. Traditional feature-based machine learning (ML) models, deep convolutional neural networks (DCNN) models and a fusion model that integrated the prediction outcomes from all models were used to classify benign and malignant TR4 nodules. A fivefold cross-validation approach was employed, and the diagnostic performance of each model and radiologists was compared.

RESULTS: In the external test data set, the area under the receiver operating characteristic curve (AUROC) of the three DCNN-based secondary transfer learning models-InceptionV3, DenseNet121 and ResNet50-were 0.852, 0.837 and 0.856, respectively. These values were higher than those of the three traditional ML models-logistic regression, multilayer perceptron and random forest-at 0.782, 0.790, and 0.767, respectively, and higher than that of an experienced radiologist (0.815). The fusion diagnostic model we developed, with an AUROC of 0.880, was found to outperform the experienced radiologist in diagnosing TR4 nodules.

CONCLUSION: The integration of artificial intelligence algorithms into medical imaging studies could improve the accuracy of identifying high-risk TR4 nodules pre-operatively and have significant clinical application potential.

来源:Ultrasound Med Biol,查看原文。版权归原作者所有。

引用

  1. Lai M, Feng B, Yao J, et al. Value of Artificial Intelligence in Improving the Accuracy of Diagnosing TI-RADS Category 4 Nodules. Ultrasound Med Biol. 2023. PMID: 37652837. doi:10.1016/j.ultrasmedbio.2023.08.008

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。