文献速递头颈部肿瘤放疗放疗物理与技术

深度学习勾画提升头颈部危及器官自动 delineation 表现

Improving automatic delineation for head and neck organs at risk by Deep Learning Contouring.
PubMed 文献 · Radiother Oncol · 2019-10-22

本研究在589例头颈癌患者上训练深度学习勾画(DLC)网络,并在104例独立验证队列中与基于图谱的勾画(ABAS)比较22个危及器官(OAR)的自动勾画性能。DLC在19/22个OAR上定量指标等于或显著优于ABAS,主观评价更受偏好,且缩短了 inexperienced observer 的勾画时间,但手动勾画仍优于两种自动方法。

要点

  • DLC在19/22个OAR上DICE等于或显著优于ABAS,改善主要见于腺体和上消化道OAR。
  • DLC整体主观评价更常优于ABAS,被认为更精确且更常与手动勾画混淆。
  • DLC显著缩短了 inexperienced observer 的勾画时间,但手动勾画仍优于DLC和ABAS。

核心结论

  • 在大型头颈癌患者队列上训练的DLC在多数头颈OAR上优于ABAS。

研究设计与样本

  • DLC神经网络在589例头颈癌患者上训练,并与ABAS在104例独立验证队列(均经手动勾画)中比较。
  • 评估22个OAR(腺体、上消化道和中枢神经系统相关结构)的DICE、|Δmean-dose|和|Δmax-dose|。
  • 对7个OAR子集评估勾画时间、观察者间变异和主观判断。

主要发现

  • DLC在19/22个OAR上定量性能指标等于或显著优于ABAS(DICE/|Δmean dose|/|Δmax dose|:0.59/4.2/4.1 Gy(ABAS);0.74/1.1/0.8 Gy(DLC))。
  • 改善主要见于腺体和上消化道OAR。
  • DLC显著缩短了 inexperienced observer 的勾画时间。
  • 主观评价显示DLC轮廓整体更常优于ABAS轮廓,被认为更精确,且更常与手动轮廓混淆。
  • 手动轮廓仍优于DLC和ABAS,但DLC结果处于或接近本队列手动编辑轮廓的观察者间变异范围。

关键数值与推荐

  • DICE/|Δmean dose|/|Δmax dose|:0.59/4.2/4.1 Gy(ABAS);0.74/1.1/0.8 Gy(DLC)。

临床怎么用

  • DLC可作为头颈OAR自动勾画的改进工具,尤其对腺体和上消化道结构,但仍需人工审核编辑。

局限性

  • 手动轮廓仍优于DLC和ABAS;DLC结果处于或接近手动编辑轮廓的观察者间变异范围。

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)。

原文

INTRODUCTION: Adequate head and neck (HN) organ-at-risk (OAR) delineation is crucial for HN radiotherapy and for investigating the relationships between radiation dose to OARs and radiation-induced side effects. The automatic contouring algorithms that are currently in clinical use, such as atlas-based contouring (ABAS), leave room for improvement. The aim of this study was to use a comprehensive evaluation methodology to investigate the performance of HN OAR auto-contouring when using deep learning contouring (DLC), compared to ABAS. METHODS: The DLC neural network was trained on 589 HN cancer patients. DLC was compared to ABAS by providing each method with an independent validation cohort of 104 patients, which had also been manually contoured. For each of the 22 OAR contours - glandular, upper digestive tract and central nervous system (CNS)-related structures - the dice similarity coefficient (DICE), and absolute mean and max dose differences (|Δmean-dose| and |Δmax-dose|) performance measures were obtained. For a subset of 7 OARs, an evaluation of contouring time, inter-observer variation and subjective judgement was performed. RESULTS: DLC resulted in equal or significantly improved quantitative performance measures in 19 out of 22 OARs, compared to the ABAS (DICE/|Δmean dose|/|Δmax dose|: 0.59/4.2/4.1 Gy (ABAS); 0.74/1.1/0.8 Gy (DLC)). The improvements were mainly for the glandular and upper digestive tract OARs. DLC significantly reduced the delineation time for the inexperienced observer. The subjective evaluation showed that DLC contours were more often preferable to the ABAS contours overall, were considered to be more precise, and more often confused with manual contours. Manual contours still outperformed both DLC and ABAS; however, DLC results were within or bordering the inter-observer variability for the manual edited contours in this cohort. CONCLUSION: The DLC, trained on a large HN cancer patient cohort, outperformed the ABAS for the majority of HN OARs.

来源:Radiother Oncol,查看原文。版权归原作者所有。

引用

  1. van Dijk LV, Van den Bosch L, Aljabar P, et al. Improving automatic delineation for head and neck organs at risk by Deep Learning Contouring. Radiother Oncol. 2019. PMID: 31653573. doi:10.1016/j.radonc.2019.09.022

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。