基于深度学习的心脏MRI规划:从定位像到电影视图使用地标检测
这项回顾性研究分析了1023例(8-90岁)心脏MRI患者的数据,开发全自动深度学习框架用于心脏MRI规划。模型在5折交叉验证中评估,各平面地标定位中位距离5.1-7.5 mm,短轴规划角度偏差2.0°(2CH)和1.5°(4CH),长轴视图使用SAX中间层角度误差低于SAX基底层。结论认为基于深度学习的自动化心脏MRI规划工作流可行且精度有所提高。
要点
- 定位像上地标定位中位距离:冠状面5.1 mm(上)和7.2 mm(下),矢状面5.6 mm(上)和7.5 mm(下)。
- 短轴中间层基于左室中心、右室插入点和右室钝角的地标中位误差5.2 mm,基底层瓣膜相关标注误差4.6 mm。
- 长轴视图角度误差:SAX中间层规划2CH为3.3°、4CH为2.6°,低于SAX基底层的4.0°和3.9°。
核心结论
- 基于深度学习的自动化心脏MRI规划工作流可行,精度有所提高。
研究设计与样本
- 回顾性研究,纳入1023例(8-90岁)接受心脏MRI的患者,使用冠状、矢状、轴位定位像及短轴和长轴电影图像;专家手动标注地标作为参考标准,采用5折交叉验证评估深度学习模型。
关键影像表现/征象
- 定位像地标定位中位距离:冠状面5.1 mm(上)和7.2 mm(下),矢状面5.6 mm(上)和7.5 mm(下)。
- 轴位、2腔和4腔地标的中位距离分别为5.2 mm、5.2 mm和5.6 mm。
- 短轴中间层基于左室中心、右室插入点和右室钝角的标注中位误差5.2 mm,基底层瓣膜相关标注误差4.6 mm。
- 短轴规划角度偏差:2CH为2.0°,4CH为1.5°。
- 长轴视图角度误差:SAX中间层规划2CH为3.3°、4CH为2.6°,SAX基底层规划2CH为4.0°、4CH为3.9°。
诊断标准与关键数值
- 定位像地标定位中位距离:冠状面上5.1 mm、下7.2 mm;矢状面上5.6 mm、下7.5 mm。
- 轴位、2腔、4腔地标中位距离分别为5.2 mm、5.2 mm、5.6 mm。
- 短轴中间层左室中心、右室插入点、右室钝角标注中位误差5.2 mm;基底层瓣膜相关标注中位误差4.6 mm。
- 短轴规划角度偏差:2CH为2.0°,4CH为1.5°。
- 长轴视图角度误差:SAX中间层2CH为3.3°、4CH为2.6°;SAX基底层2CH为4.0°、4CH为3.9°。
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
RATIONALE AND OBJECTIVES: This study evaluates a fully automated deep learning framework to enhance the efficiency and accuracy of cardiac MRI planning.
MATERIALS AND METHODS: In this retrospective study, data from 1023 patients (ages 8-90 years) who underwent cardiac MRI were analyzed, including coronal, sagittal, axial localizers, and short-axis (SAX) and long-axis cine images. Experts manually annotated landmarks, serving as the ground truth for developing deep learning models. The models were assessed using 5-fold cross-validation. Performance metrics included median landmark distances and plane angle differences.
RESULTS: The model achieved robust performance in landmark localization across all cardiac MRI planes. For localizer images, median distances were 5.1 mm (superior) and 7.2 mm (inferior) on coronal views, and 5.6 mm (superior) and 7.5 mm (inferior) on sagittal views. Median distances for axial, 2-chamber, and 4-chamber landmarks were 5.2 mm, 5.2 mm, and 5.6 mm, respectively. In short-axis mid slices, annotations based on the left ventricular center, right ventricular insertion points, and right ventricle obtuse angle had a median error of 5.2 mm, while basal slice valve-based annotations had 4.6 mm error. Angular deviations for SAX planning were 2.0° (2CH) and 1.5° (4CH). For long-axis views, angulation errors were lower using SAX mid slices (3.3° for 2CH, 2.6° for 4CH) compared to SAX base (4.0° and 3.9°, respectively).
CONCLUSION: A deep learning-based automated workflow for cardiac MRI planning is feasible with improved precision.
引用
- Dhruba DD, Goetz S, Pria OFD, et al. Deep Learning-Based Cardiac MRI Planning from Localizers to Cine Views Using Landmark Detection. Acad Radiol. 2025. PMID: 41353071. doi:10.1016/j.acra.2025.11.028
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。