文献速递介入技术与安全CT

增强现实配准方法用于图像引导介入的多中心评估

Multicenter assessment of augmented reality registration methods for image-guided interventions.
PubMed 文献 · Radiol Med · 2022-06-23

在两个学术中心,18名不同培训阶段的参与者使用HoloLens 2将3D CT体模模型配准到CT网格,比较手动与自动配准时间及精度,并回顾性对比HoloLens 1。单手手势、双手手势和Xbox控制器平均手动配准时间分别为27.7秒、24.3秒和72.8秒,自动配准为5.3秒(ANOVA p < 0.0001);不同经验水平间无显著差异(p > 0.05);手势配准在HoloLens 2上较HoloLens 1快81.7%(p < 0.05)。

要点

  • HoloLens 2自动配准平均5.3秒,单手手势27.7秒,双手手势24.3秒,Xbox控制器72.8秒。
  • 不同经验水平(主治、住院医、研究员、医学生)间配准时间无显著差异(p > 0.05)。
  • 手势配准在HoloLens 2上较HoloLens 1快81.7%(p < 0.05)。

核心结论

  • HoloLens 2手动配准时间显著缩短,接近自动配准并优于Xbox控制器配准。

研究设计与样本

  • 多中心研究,两个学术中心,18名处于不同临床培训阶段的参与者,使用HoloLens 2将3D CT体模模型配准到CT网格,并回顾性对比既往HoloLens 1结果。

诊断标准与关键数值

  • 单手手势、双手手势和Xbox控制器平均手动配准时间分别为27.7秒、24.3秒和72.8秒,自动配准平均5.3秒(ANOVA p < 0.0001);配准精度分别为5.9毫米、9.5毫米和8.6毫米,手动配准方法间无显著差异(p > 0.05)。

临床应用与阅片要点

  • 快速、自适应且准确的横断面影像全息模型配准对于在CT引导介入中实施增强现实辅助3D导航至关重要。

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 manual and automatic registration times and registration accuracies on HoloLens 2 for aligning a 3D CT phantom model onto a CT grid, a crucial step for intuitive 3D navigation during CT-guided interventions; to compare registration times between HoloLens 1 and 2.

METHODS: Eighteen participants in various stages of clinical training across two academic centers performed registration of a 3D CT phantom model onto a CT grid using HoloLens 2. Registration times and accuracies were compared among different registration methods, clinical experience levels, and consecutive attempts. Registration times were also compared retrospectively to prior HoloLens 1 results.

RESULTS: Mean aggregate manual registration times were 27.7 s, 24.3 s, and 72.8 s for one-handed gesture, two-handed gesture, and Xbox controller, respectively; mean automatic registration time was 5.3 s (ANOVA p < 0.0001). No significant difference in registration times was found among attendings, residents and fellows, and medical students (p > 0.05). Significant improvements in registration times were detected across consecutive attempts using hand gestures (p < 0.01). Compared to prior HoloLens 1 data, hand gesture registration was 81.7% faster with HoloLens 2 (p < 0.05). Registration accuracies were not significantly different across manual registration methods, measuring at 5.9 mm, 9.5 mm, and 8.6 mm with one-handed gesture, two-handed gesture, and Xbox controller, respectively (p > 0.05).

CONCLUSIONS: Manual registration times decreased significantly on HoloLens 2, approaching those of automatic registration and outperforming Xbox controller registration. Fast, adaptive, and accurate registration of holographic models of cross-sectional imaging is paramount for the implementation of augmented reality-assisted 3D navigation during CT-guided interventions.

来源:Radiol Med,查看原文。版权归原作者所有。

引用

  1. Li, Wakim, Koethe, et al. Multicenter assessment of augmented reality registration methods for image-guided interventions. Radiol Med. 2022. PMID: 35737194. doi:10.1007/s11547-022-01515-3

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。