文献速递腹部CT

基于多尺度深度强化学习的全自动肝脏分割算法验证及其与手动分割的比较

Validation of a fully automated liver segmentation algorithm using multi-scale deep reinforcement learning and comparison versus manual segmentation.
PubMed 文献 · Eur J Radiol · 2020-03-05

回顾性纳入462例多期相CT共2772个序列,评估基于多尺度深度强化学习的全自动肝脏容积分析软件,并与手动分割比较。自动测量平均绝对误差44.3 mL(占平均肝容积2.37%),肝容积不依赖对比期相(p=0.697)或层厚(p=0.446),平均处理时间9.94秒 vs 手动219.34秒,ICC 0.996,结论为全自动肝脏容积分析准确、可重复、稳健、快速且与手动分割高度一致。

要点

  • 自动肝脏容积测量平均绝对误差44.3 mL,占平均肝容积2.37%。
  • 肝容积测量不受对比期相(p=0.697)和层厚(p=0.446)影响。
  • 自动处理每数据集9.94秒,手动分割219.34秒,ICC 0.996。

核心结论

  • 基于多尺度深度强化学习的全自动肝脏容积分析准确、可重复、稳健、快速,与手动分割高度一致。

研究设计与样本

  • 回顾性获得462例多期相CT数据集,每例6个序列(3个不同对比期相和1.5/5 mm两种层厚重建),共2772个序列。
  • AI算法采用多尺度深度强化学习进行3D体部标记检测和3D结构分割,约5000个数据集用于肝脏容积测量训练。
  • 60个数据集由3位放射科医生手动分割,其中20个数据集每位医生分割3次,用于与自动结果和参考标准比较。

诊断标准与关键数值

  • 自动测量平均绝对误差44.3 mL,占平均肝容积的2.37%。
  • 肝容积不依赖对比期相(p=0.697)和层厚(p=0.446)。
  • 自动与手动方法一致性ICC为0.996。

临床应用与阅片要点

  • 原文未据此给出推荐。

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

PURPOSE: To evaluate the performance of an artificial intelligence (AI) based software solution tested on liver volumetric analyses and to compare the results to the manual contour segmentation.

MATERIALS AND METHODS: We retrospectively obtained 462 multiphasic CT datasets with six series for each patient: three different contrast phases and two slice thickness reconstructions (1.5/5 mm), totaling 2772 series. AI-based liver volumes were determined using multi-scale deep-reinforcement learning for 3D body markers detection and 3D structure segmentation. The algorithm was trained for liver volumetry on approximately 5000 datasets. We computed the absolute error of each automatically- and manually-derived volume relative to the mean manual volume. The mean processing time/dataset and method was recorded. Variations of liver volumes were compared using univariate generalized linear model analyses. A subgroup of 60 datasets was manually segmented by three radiologists, with a further subgroup of 20 segmented three times by each, to compare the automatically-derived results with the ground-truth.

RESULTS: The mean absolute error of the automatically-derived measurement was 44.3 mL (representing 2.37 % of the averaged liver volumes). The liver volume was neither dependent on the contrast phase (p = 0.697), nor on the slice thickness (p = 0.446). The mean processing time/dataset with the algorithm was 9.94 s (sec) compared to manual segmentation with 219.34 s. We found an excellent agreement between both approaches with an ICC value of 0.996.

CONCLUSION: The results of our study demonstrate that AI-powered fully automated liver volumetric analyses can be done with excellent accuracy, reproducibility, robustness, speed and agreement with the manual segmentation.

来源:Eur J Radiol,查看原文。版权归原作者所有。

引用

  1. Winkel DJ, Weikert TJ, Breit HC, et al. Validation of a fully automated liver segmentation algorithm using multi-scale deep reinforcement learning and comparison versus manual segmentation. Eur J Radiol. 2020. PMID: 32171914. doi:10.1016/j.ejrad.2020.108918

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。