文献速递妇产影像泌尿生殖技术与安全US

基于深度卷积神经网络的卵巢附件病变超声图像O-RADS自动分类研究

A Study on Automatic O-RADS Classification of Sonograms of Ovarian Adnexal Lesions Based on Deep Convolutional Neural Networks.
PubMed 文献 · Ultrasound Med Biol · 2024-11-26

本研究基于深度卷积神经网络探索卵巢附件病变超声图像O-RADS自动分类新方法,纳入2455幅发育数据集图像和426幅时间验证数据集图像,由三名资深超声医师按O-RADS v2022分类;ConvNeXt-Tiny模型在发育数据集准确率0.878、验证数据集0.859,模型辅助显著缩短三名医师分类时间(Cohen's d=5.75)。

要点

  • 恶性病变(O-RADS 4类和5类)在发育数据集敏感度0.949、特异度0.892,验证数据集敏感度0.962、特异度0.842。
  • ConvNeXt-Tiny在发育数据集准确率0.878,验证数据集0.859,测试集间无显著差异。
  • 模型辅助显著缩短三名不同年资超声医师的O-RADS分类时间(Cohen's d=5.75)。

核心结论

  • ConvNeXt-Tiny在O-RADS 2-5类分类中表现出稳健且稳定的性能,提高了超声医师的分类效率。

研究设计与样本

  • 回顾性研究,发育数据集含870个卵巢附件病变的2455幅二维灰度超声图像,时间验证数据集含280个病变的426幅图像,由三名资深超声医师按O-RADS v2022分类,开发ResNet34、DenseNet121和ConvNeXt-Tiny三种深度卷积神经网络模型并采用迁移学习技术。

诊断标准与关键数值

  • 恶性病变(O-RADS 4类和5类)在发育数据集敏感度0.949、特异度0.892。
  • 恶性病变(O-RADS 4类和5类)在时间验证数据集敏感度0.962、特异度0.842。
  • ResNet34、DenseNet121和ConvNeXt-Tiny在发育数据集超声图像预测的总体准确率分别为0.737、0.752和0.878。
  • ConvNeXt-Tiny模型在时间验证数据集超声图像预测的准确率为0.859,测试集间无显著差异。

临床应用与阅片要点

  • 模型辅助可显著缩短三名不同年资超声医师的O-RADS分类时间(Cohen's d=5.75)。

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: This study explored a new method for automatic O-RADS classification of sonograms based on a deep convolutional neural network (DCNN).

METHODS: A development dataset (DD) of 2,455 2D grayscale sonograms of 870 ovarian adnexal lesions and an intertemporal validation dataset (IVD) of 426 sonograms of 280 lesions were collected and classified according to O-RADS v2022 (categories 2-5) by three senior sonographers. Classification results verified by a two-tailed z-test to be consistent with the O-RADS v2022 malignancy rate indicated the diagnostic performance was comparable to that of a previous study and were used for training; otherwise, the classification was repeated by two different sonographers. The DD was used to develop three DCNN models (ResNet34, DenseNet121, and ConvNeXt-Tiny) that employed transfer learning techniques. Model performance was assessed for accuracy, precision, and F1 score, among others. The optimal model was selected and validated over time using the IVD and to analyze whether the efficiency of O-RADS classification was improved with the assistance of this model for three sonographers with different years of experience.

RESULTS: The proportion of malignant tumors in the DD and IVD in each O-RADS-defined risk category was verified using a two-tailed z-test. Malignant lesions (O-RADS categories 4 and 5) were diagnosed in the DD and IVD with sensitivities of 0.949 and 0.962 and specificities of 0.892 and 0.842, respectively. ResNet34, DenseNet121, and ConvNeXt-Tiny had overall accuracies of 0.737, 0.752, and 0.878, respectively, for sonogram prediction in the DD. The ConvNeXt-Tiny model's accuracy for sonogram prediction in the IVD was 0.859, with no significant difference between test sets. The modeling aid significantly reduced O-RADS classification time for three sonographers (Cohen's d = 5.75).

CONCLUSION: ConvNeXt-Tiny showed robust and stable performance in classifying O-RADS 2-5, improving sonologists' classification efficacy.

来源:Ultrasound Med Biol,查看原文。版权归原作者所有。

引用

  1. Liu T, Miao K, Tan G, et al. A Study on Automatic O-RADS Classification of Sonograms of Ovarian Adnexal Lesions Based on Deep Convolutional Neural Networks. Ultrasound Med Biol. 2024. PMID: 39603844. doi:10.1016/j.ultrasmedbio.2024.11.009

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。