基于UNet深度学习在乳腺X线摄影中量化乳腺动脉钙化以检测心血管疾病
本研究回顾性纳入369例患者的乳腺X线摄影,开发并验证一种改进U-Net深度学习模型用于检测、分割和量化乳腺动脉钙化(BAC)。模型分割Dice系数训练集0.90、验证集0.89,BAC检测F1分数验证集0.97、训练集0.93,Bland-Altman分析训练集平均钙量差异-0.98 mg。作者认为该框架可提升BAC检测、分割与量化,支持女性医疗中整合双重目的筛查的潜力。
要点
- 改进U-Net模型在乳腺X线摄影上分割BAC的Dice系数为训练集0.90、验证集0.89。
- BAC检测F1分数为验证集0.97、训练集0.93。
- 量化可靠性经Bland-Altman分析,训练集平均钙量差异为-0.98 mg。
核心结论
- 该深度学习框架大幅改善乳腺X线摄影中BAC的检测、分割与量化,提升心血管风险筛查的准确性与效率,支持女性医疗中整合双重目的筛查的潜力。
研究设计与样本
- 回顾性研究,使用369例患者的乳腺X线摄影;采用改进U-Net架构,结合Hausdorff损失、Dice损失和二元交叉熵(BCE)损失进行分割与后续量化;以Dice分数、分割BCE损失、线性拟合及Bland-Altman分析评估模型性能。
诊断标准与关键数值
- 分割准确度:Dice分数训练集0.90、验证集0.89。
- BAC检测分类准确度:F1分数验证集0.97、训练集0.93。
- 量化可靠性:Bland-Altman分析显示训练集平均钙量差异为-0.98 mg。
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: Breast arterial calcification (BAC) is increasingly recognized as a significant indicator of cardiovascular risk, necessitating improvements in detection and quantification methods through mammographic screening.
PURPOSE: To develop and validate a deep-learning model capable of detecting, segmenting, and quantifying BAC in mammograms, improving mammographic screening for cardiovascular risk assessment.
MATERIALS AND METHODS: We conducted a retrospective study using mammograms from 369 patients. The study utilized a modified U-Net architecture that incorporates Hausdorff loss, Dice loss, and Binary Cross-Entropy (BCE) loss for segmentation and subsequent quantification. The model's performance was assessed using the Dice score, BCE loss for segmentation accuracy, linear fit, and Bland-Altman analysis for quantification accuracy.
RESULTS: Our model achieved high segmentation accuracy with Dice scores of 0.90 for the training set and 0.89 for the validation set. Quantification reliability was validated through Bland-Altman analysis, showing a mean difference of -0.98 mg of calcium in the training set. The model also demonstrated high classification accuracy with F1 scores of 0.97 and 0.93 for validation and training sets, respectively, in BAC detection.
CONCLUSION: The deep-learning framework substantially improves BAC detection, segmentation, and quantification in mammograms, advancing the accuracy and efficiency of cardiovascular risk screening. This study supports the potential for integrated dual-purpose screening in women's healthcare.
引用
- Li W, Zhang Q, Black D, et al. Quantification of Breast Arterial Calcification in Mammograms Using a UNet-Based Deep Learning for Detecting Cardiovascular Disease. Acad Radiol. 2025. PMID: 40541546. doi:10.1016/j.acra.2025.05.036
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。