文献速递神经技术与安全CT

混合精度训练深度学习在脑出血CT分割中的粗到细策略研究

Deep-learning measurement of intracerebral haemorrhage with mixed precision training: a coarse-to-fine study.
PubMed 文献 · Clin Radiol · 2023-01-22

本研究开发了一种统一的深度学习方法,用于不同层厚CT图像上的脑出血自动分割,采用粗到细策略包括三维颅骨剥离、三维血肿定位和二维精细分割,在内部和外部数据集上平均Dice系数达0.887,自动分割时间约20秒,比手动分割快5-42倍。

要点

  • 粗到细深度学习策略在脑出血CT分割中平均Dice系数为0.887。
  • 自动分割在不同层厚(0.625/1.25/5 mm)下体积一致性良好(p<0.05)。
  • 自动分割时间稳定在20.01±2.03秒,比手动分割快5-42倍。

核心结论

  • 粗到细深度学习策略在CT图像上实现了最佳的脑出血分割性能。
  • 自动分割在不同层厚和血肿范围下均保持较一致的体积测量和分割时间。

研究设计与样本

  • 研究纳入134例患者,来自内部数据库(67例)和外部数据库CQ500(67例)。
  • CT检查包括0.625、1.25和5 mm等多种层厚。
  • 分割采用三阶段粗到细策略:三维颅骨剥离分割、三维脑出血定位分割和二维脑出血精细分割。
  • 三个阶段共享同一sICHNet分割网络,并采用混合精度训练加速训练过程。
  • 二维精细分割中将连续切片格式化为三通道图像以保留三维上下文信息。

关键影像表现/征象

  • 粗到细分割策略取得了最佳性能,平均Dice系数为0.887。

诊断标准与关键数值

  • 平均Dice系数0.887用于评估脑出血分割性能。
  • 手动与自动分割之间、以及同一患者不同层厚分割之间脑出血体积一致性p<0.05。
  • 自动分割时间20.01±2.03秒,不受CT层厚和出血范围影响。

临床应用与阅片要点

  • 该自动分割方法可用于不同层厚CT图像的脑出血体积测量,有助于纵向随访比较。
  • 混合精度训练和粗到细策略可加速深度学习模型训练和推理,适合临床工作流。
  • 研究还可视化展示了保守治疗和手术治疗的纵向研究,提示可用于疗效评估。

局限性

  • 样本量较小(共134例),内部和外部数据库各67例。
  • 未报告敏感度、特异度、AUC等诊断性能指标。
  • 研究为回顾性设计,未提及前瞻性验证。

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

AIM: To develop a unified deep-learning-based method for automated intracerebral haemorrhage (ICH) segmentation on computed tomography (CT) images with different layer thickness parameters.

MATERIALS AND METHODS: A total of 134 patients from an internal database (67 patients) and an external database (called CQ500, 67 patients) were employed. The CT examinations included multiple layer thicknesses such as 0.625, 1.25 and 5 mm. ICH segmentation was performed by a coarse-to-fine strategy, including three stages of three-dimensional (3D) skull-stripping segmentation, 3D ICH localisation segmentation, and two-dimensional (2D) ICH fine segmentation. The three stages shared the same sICHNet for segmentation and employed mixed precision training to speed up the training process. In addition, the 3D contextual information from CT was maintained by formatting the consecutive slices into a three-channel image in the 2D ICH fine segmentation.

RESULTS: Experimental results demonstrated that the coarse-to-fine segmentation strategy achieved the best performance with a mean Dice coefficient of 0.887. ICH volume consistency was observed (p<0.05) between manual and automatic segmentations, and between segmentations of same individual but with different layer thicknesses in internal dataset and external database. Automated segmentation achieved a relatively consistent segmentation time of 20.01 ± 2.03 seconds no matter the layer thickness of the CT images and the extent of ICH. Longitudinal studies with conservative management and surgical treatment were also visualised.

CONCLUSIONS: The coarse-to-fine deep learning strategy achieved the best ICH segmentation performance on CT images. The automated segmentation was 5-42 times faster than manual segmentation given ICH of different extents and using different layer thickness parameters.

来源:Clin Radiol,查看原文。版权归原作者所有。

引用

  1. Jiang, Wang, Zheng, et al. Deep-learning measurement of intracerebral haemorrhage with mixed precision training: a coarse-to-fine study. Clin Radiol. 2023. PMID: 36746725. doi:10.1016/j.crad.2022.12.019

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。