文献速递胸部心血管技术与安全US

AutoRV:机械通气患者经食管超声自动评估右心室结构与功能

AutoRV: Automated Assessment of Right Ventricular Structure and Function Using Transesophageal Echocardiography in Mechanically Ventilated Patients.
PubMed 文献 · Ultrasound Med Biol · 2026-06-19

本研究基于51例机械通气患者的二维经食管超声图像,开发AutoRV全自动流程检测右心室心尖与三尖瓣环标志点并计算功能参数。U-Net处理速度达228帧/秒,标志点定位误差为4.2±3.4mm。自动测量与人工测量的一致性接近两名临床医师间的人工测量变异性。

要点

  • AutoRV可自动计算TAPSE、FAC及线性应变,有望用于ICU右心室连续监测。
  • 自动与人工测量的一致性与医师间人工测量变异性相当。
  • U-Net速度快且精度可接受,适合实时经食管超声右心室监测。

核心结论

  • AutoRV全自动端到端方案准确度可接受且时间效率极高,有潜力用于ICU患者右心室自动监测。

研究设计与样本

  • 研究类型为算法开发与验证,纳入51例患者的二维经食管超声图像进行训练和测试。
  • 使用多个深度卷积神经网络检测右心室心尖和三尖瓣环标志点,并与有经验临床医师的人工测量比较。
  • 基于心尖及三尖瓣环标志点计算三尖瓣环收缩期位移、右心室面积变化分数和线性右心室应变。

关键影像表现/征象

  • U-Net架构以228帧/秒速度处理图像,标志点检测距离对应真实值的误差为4.2±3.4mm。
  • 三尖瓣环收缩期位移误差为1.1±2.3mm。
  • 右心室面积变化分数误差为0.2±6.6%。
  • 线性右心室应变的误差为3.9±3.6%。

诊断标准与关键数值

  • U-Net标志点定位误差为4.2±3.4mm,处理速度为228帧/秒。
  • 三尖瓣环收缩期位移自动测量误差为1.1±2.3mm。
  • 右心室面积变化分数自动测量误差为0.2±6.6%。
  • 线性右心室应变自动测量误差为3.9±3.6%。
  • 自动方案与人工测量的吻合度与两名临床医师间人工测量变异性相当。

临床应用与阅片要点

  • AutoRV可从二维经食管超声自动检测右心室心尖和三尖瓣环标志点,减少人工勾画时间。
  • 可自动输出TAPSE、FAC和线性应变,供ICU机械通气患者右心室功能监测参考。
  • 阅片或报告时需注意自动测量误差约为毫米级,TAPSE误差1.1±2.3mm,应变误差3.9±3.6%。

局限性

  • 样本仅来源于51例患者的图像,样本量较小。
  • 原文比较对象为有经验临床医师,缺乏外部验证或与其他超声设备的对照。

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: Right ventricle (RV) dysfunction has therapeutic implications for the management of mechanically ventilated patients in intensive care units. However, RV function is usually qualitatively evaluated by clinicians, as quantitative manual assessment is time-consuming and imprecise. Therefore, we developed AutoRV, an automated monitoring pipeline to assess RV function from 2-D transesophageal echocardiography using deep learning-based landmark detection.

METHODS: Multiple deep convolutional neural networks (CNNs) were implemented to automatically detect the apex of the RV and the tricuspid annulus points in 2-D transesophageal echocardiography images. All networks were trained and tested on images acquired from 51 patients. Multiple parameters were computed based on the apex and tricuspid annulus landmarks, including tricuspid annular plane systolic excursion, RV fractional area change and linear RV strain (RVSglobal). Automated quantifications were compared with manual measurements performed by an experienced clinician.

RESULTS: The U-Net architecture yielded the most promising results-processing images at 228 frames/s and detecting landmarks at a distance of 4.2±3.4mm from the corresponding ground truth. The resulting errors in tricuspid annular plane systolic excursion, RV fractional area change and RVSglobal were equal to 1.1±2.3mm,0.2±6.6% and 3.9±3.6%, respectively. The agreement between our automatic solution and manual measurements paralleled the inter-observer variability of manual measurements between two clinicians.

CONCLUSION: Our fully automatic end-to-end solution, AutoRV, proved to be acceptably accurate and extremely time-efficient, suggesting its potential for automating RV monitoring in intensive care unit patients.

来源:Ultrasound Med Biol,查看原文。版权归原作者所有。

引用

  1. Missana, Hau, Grude, et al. AutoRV: Automated Assessment of Right Ventricular Structure and Function Using Transesophageal Echocardiography in Mechanically Ventilated Patients. Ultrasound Med Biol. 2026. PMID: 42321092. doi:10.1016/j.ultrasmedbio.2026.05.002

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。