乳腺X线摄影可疑钙化中纹理分析与机器学习联合应用:有望避免不必要的立体定位活检
回顾性分析235例乳腺X线摄影BI-RADS 4类可疑钙化、接受真空辅助活检患者的放大位图像,手动分割微钙化后提取249个纹理特征,经主成分分析降维后训练多层感知器神经网络,以组织学为参考标准。最终226例(良性150例、恶性76例)纳入分析,测试集(n=54)AUC为0.82(95%CI 0.70-0.94)与0.832(95%CI 0.72-0.94),高敏感度阈值标准可避免37.1-45.7%的不必要活检,代价为每位阅片者1例假阴性。
要点
- 纹理分析联合机器学习对乳腺X线摄影可疑钙化的测试集AUC为0.82(95%CI 0.70-0.94)。
- 高敏感度阈值在测试集可避免37.1-45.7%的不必要活检,代价为每位阅片者1例假阴性。
- 该方法可能用于可疑乳腺钙化的风险分层,但为回顾性探索性研究。
核心结论
- 联合纹理分析与机器学习可用于乳腺X线摄影可疑钙化的风险分层,以较低的假阴性代价避免不必要的活检。
研究设计与样本
- 回顾性分析两年期间235例因可疑钙化(BI-RADS 4类)行真空辅助活检患者的乳腺X线摄影放大位图像,使用纹理分析工具MaZda(版本4.6)。
- 由两位阅片者手动分割微钙化,提取灰度直方图、灰度共生矩阵和游程矩阵共249个图像特征。
- 经主成分分析降维后,以组织学结果为参考标准训练多层感知器人工神经网络,数据集按70%与30%分为训练集与测试集。
- 因9例数据缺失,最终226例(良性150例、恶性76例)纳入分析,测试集n=54。
诊断标准与关键数值
- 特征选择后得到9个图像特征用于多层感知器训练。
- 测试集(n=54)ROC曲线下面积两位阅片者分别为0.82(95%CI 0.70-0.94)和0.832(95%CI 0.72-0.94)。
- 在训练集确定的高敏感度阈值标准成功应用于测试集,可避免37.1-45.7%的不必要活检,代价为每位阅片者1例假阴性。
临床应用与阅片要点
- 作者认为联合纹理分析与机器学习可用于可疑乳腺钙化的风险分层,以较低的假阴性代价避免不必要的活检。
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
OBJECTIVES: To investigate whether combined texture analysis and machine learning can distinguish malignant from benign suspicious mammographic calcifications, to find an exploratory rule-out criterion to potentially avoid unnecessary benign biopsies.
METHODS: Magnification views of 235 patients which underwent vacuum-assisted biopsy of suspicious calcifications (BI-RADS 4) during a two-year period were retrospectively analyzed using the texture analysis tool MaZda (Version 4.6). Microcalcifications were manually segmented and analyzed by two readers, resulting in 249 image features from gray-value histogram, gray-level co-occurrence and run-length matrices. After feature reduction with principal component analysis (PCA), a multilayer perceptron (MLP) artificial neural network was trained using histological results as the reference standard. For training and testing of this model, the dataset was split into 70 % and 30 %. ROC analysis was used to calculate diagnostic performance indices.
RESULTS: 226 patients (150 benign, 76 malignant) were included in the final analysis due to missing data in 9 cases. Feature selection yielded nine image features for MLP training. Area under the ROC-curve in the testing dataset (n = 54) was 0.82 (95 %-CI: 0.70-0.94) and 0.832 (95 %-CI 0.72-0.94) for both readers, respectively. A high sensitivity threshold criterion was identified in the training dataset and successfully applied to the testing dataset, demonstrating the potential to avoid 37.1-45.7 % of unnecessary biopsies at the cost of one false-negative for each reader.
CONCLUSION: Combined texture analysis and machine learning could be used for risk stratification in suspicious mammographic calcifications. At low costs in terms of false-negatives, unnecessary biopsies could be avoided.
引用
- Stelzer PD, Steding O, Raudner MW, et al. Combined texture analysis and machine learning in suspicious calcifications detected by mammography: Potential to avoid unnecessary stereotactical biopsies. Eur J Radiol. 2020. PMID: 33010682. doi:10.1016/j.ejrad.2020.109309
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。