文献速递乳腺甲状腺与浅表技术与安全US

深度学习鉴别超声弹性成像筛查发现的乳腺肿块

Deep Learning for Differentiation of Breast Masses Detected by Screening Ultrasound Elastography.
PubMed 文献 · Ultrasound Med Biol · 2023-01-19

该回顾性研究纳入239例连续患者的245个乳腺肿块(146个良性,99个恶性)的弹性成像图像,构建基于GoogLeNet的CNN模型预测恶性概率,并与脂肪-病灶比(FLR)和5点视觉颜色弹性评分比较。CNN的AUC为0.895,显著高于FLR的0.693和弹性评分的0.641(p < 0.001),且阳性预测值从57%-78%提高至94%。

要点

  • CNN模型鉴别乳腺肿块良恶性的AUC为0.895,显著优于FLR和弹性评分。
  • CNN的阳性预测值达0.941,可能减少不必要的活检推荐。
  • 弹性评分敏感度仅0.350,需注意假阴性风险。

核心结论

  • 基于CNN的深度学习模型对超声弹性成像筛查发现的乳腺肿块良恶性鉴别诊断性能优于FLR和弹性评分,并可提高阳性预测值。

研究设计与样本

  • 回顾性研究,纳入239例连续患者的245个乳腺肿块(146个良性,99个恶性)的弹性成像图像。
  • 数据集随机分为训练集(55%)、验证集(25%)和测试集(20%)。
  • 使用GoogLeNet架构(ImageNet预训练)构建预测恶性概率的深度学习模型,训练50个epoch。
  • 参考标准未在原文摘要中明确说明。

诊断标准与关键数值

  • CNN模型敏感度0.800,特异度0.966,准确率0.898,阳性预测值0.941,阴性预测值0.875,AUC 0.895。
  • FLR敏感度0.800,特异度0.586,准确率0.673,阳性预测值0.571,阴性预测值0.810,AUC 0.693。
  • 弹性评分(5点视觉颜色评估)敏感度0.350,特异度0.931,准确率0.694,阳性预测值0.778,阴性预测值0.675,AUC 0.641。
  • CNN的AUC显著高于FLR或弹性评分(p < 0.001)。

临床应用与阅片要点

  • CNN模型可提高阳性预测值至94%,有助于减少超声筛查中乳腺肿块的不必要活检推荐。
  • 弹性评分敏感度低(0.350),阅片时需警惕假阴性,不应单独依赖弹性评分排除恶性。

局限性

  • 回顾性研究设计。
  • 单中心数据。
  • 样本量有限(245个肿块)。

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

Recently, deep learning using convolutional neural networks (CNNs) has yielded consistent results in image-pattern recognition. This study was aimed at investigating the effectiveness of deep learning using CNNs to differentiate benign and malignant breast masses identified by elastography on ultrasound screening. A data set of the elastography images of 245 breast masses (146 benign, 99 malignant) in 239 consecutive patients was retrospectively obtained. The data set was randomly split into training (55%), validation (25%) and test (20%) cohorts. A deep learning model predicting the probability of malignancy was constructed using GoogLeNet architectures (pre-trained by ImageNet) with 50 epochs. The model was then applied to the test data, and the results were compared with those obtained by evaluating the fat-to-lesion ratio (FLR) and by a 5-point visual color assessment (elasticity score). The receiver operating characteristic (ROC) curve was calculated to evaluate the performance of the model. The DeLong test was used to compare the areas under the ROC curve (AUCs). The CNN, FLR and elasticity score had a sensitivity of 0.800, 0.800 and 0.350; specificity of 0.966, 0.586 and 0.931; accuracy of 0.898, 0.673 and 0.694; positive predictive value of 0.941, 0.571 and 0.778; negative predictive value of 0.875, 0.810 and 0.675; and AUC of 0.895, 0.693 and 0.641, respectively. The AUC of the CNN was significantly higher than that of the FLR or elasticity score (p < 0.001). A CNN-based deep learning model for predicting benign or malignant breast masses revealed better diagnostic performance than did FLR or elasticity score-based estimations on ultrasound elastography. The CNN-based model also increased the positive predictive value from 57%-78% to 94%. Therefore, this model may reduce unnecessary biopsy recommendations for masses detected on breast ultrasound screening.

来源:Ultrasound Med Biol,查看原文。版权归原作者所有。

引用

  1. Fukuda, Tsunoda, Yagishita, et al. Deep Learning for Differentiation of Breast Masses Detected by Screening Ultrasound Elastography. Ultrasound Med Biol. 2023. PMID: 36681608. doi:10.1016/j.ultrasmedbio.2022.12.003

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。