文献速递消化道肿瘤放疗放疗物理与技术

深度学习在多参数MRI上自动分割胰腺肿瘤靶区

Auto-segmentation of pancreatic tumor in multi-parametric MRI using deep convolutional neural networks.
PubMed 文献 · Radiother Oncol · 2020-02-08

研究构建并验证一种基于方形窗口卷积神经网络(CNN)的算法,用于在胰腺癌多参数MRI上自动分割大体肿瘤体积(GTV)。模型在37套DCE MRI训练集上训练,在13例患者的19套DCE MRI上测试,其DSC、HD、MSD与专家间观察者差异无统计学显著差异,性能与专家放射肿瘤科医生相当。

要点

  • CNN自动分割胰腺GTV的DSC为0.73±0.09,与专家间一致性(0.71±0.08)无显著差异。
  • HD为8.11±4.09 mm、MSD为1.82±0.84 mm,与观察者间变异无显著差异。
  • 该框架有望支持MR-Linac上MRI引导在线自适应放疗的GTV快速自动分割。

核心结论

  • 基于CNN的模型可在多参数MRI上自动分割胰腺GTV,性能与专家放射肿瘤科医生相当,为MRgOART中整合多模态与每日MRI提供框架。

研究设计与样本

  • 采用含三个卷积层块的方形窗口CNN架构,训练集为27例患者共37套DCE MRI提取的约245,000个正常与230,000个肿瘤patch,并经数据增强。
  • 图像经偏置校正、强度标准化并重采样至1×1×3 mm3固定体素;测试集为另13例患者的19套DCE MRI,与放射科及放射肿瘤科专家手工分割GTV比较DSC、HD、MSD。

主要发现

  • 测试集平均DSC=0.73±0.09、HD=8.11±4.09 mm、MSD=1.82±0.84 mm。
  • 观察者间变异为DSC=0.71±0.08、HD=7.36±2.72 mm、MSD=1.78±0.66 mm,与模型性能比较的p值分别为0.6、0.52、0.88,无显著差异。

关键数值与推荐

  • 模型性能评价指标为DSC、HD与MSD;与观察者间变异比较采用p值0.6、0.52、0.88,均无显著差异。
  • 训练patch数约为245,000个正常与230,000个肿瘤patch,来自37套DCE MRI;测试为19套DCE MRI。

临床怎么用

  • 可将该CNN模型作为胰腺癌MRgOART中GTV自动分割的辅助工具,用于快速初勾并与专家手动分割比对。
  • 模型输出仍需结合专家审核,尤其关注HD所示边界偏差。

局限性

  • 摘要未报告模型在前瞻性临床流程中的实时性能与泛化验证。
  • 摘要未报告多参数序列中除DCE MRI外其他序列的单独贡献。

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

PURPOSE: The recently introduced MR-Linac enables MRI-guided Online Adaptive Radiation Therapy (MRgOART) of pancreatic cancer, for which fast and accurate segmentation of the gross tumor volume (GTV) is essential. This work aims to develop an algorithm allowing automatic segmentation of the pancreatic GTV based on multi-parametric MRI using deep neural networks. METHODS: We employed a square-window based convolutional neural network (CNN) architecture with three convolutional layer blocks. The model was trained using about 245,000 normal and 230,000 tumor patches extracted from 37 DCE MRI sets acquired in 27 patients with data augmentation. These images were bias corrected, intensity standardized, and resampled to a fixed voxel size of 1 × 1 × 3 mm3. The trained model was tested on 19 DCE MRI sets from another 13 patients, and the model-generated GTVs were compared with the manually segmented GTVs by experienced radiologist and radiation oncologists based on Dice Similarity Coefficient (DSC), Hausdorff Distance (HD), and Mean Surface Distance (MSD). RESULTS: The mean values and standard deviations of the performance metrics on the test set were DSC = 0.73 ± 0.09, HD = 8.11 ± 4.09 mm, and MSD = 1.82 ± 0.84 mm. The interobserver variations were estimated to be DSC = 0.71 ± 0.08, HD = 7.36 ± 2.72 mm, and MSD = 1.78 ± 0.66 mm, which had no significant difference with model performance at p values of 0.6, 0.52, and 0.88, respectively. CONCLUSION: We developed a CNN-based model for auto-segmentation of pancreatic GTV in multi-parametric MRI. Model performance was comparable to expert radiation oncologists. This model provides a framework to incorporate multimodality images and daily MRI for GTV auto-segmentation in MRgOART.

来源:Radiother Oncol,查看原文。版权归原作者所有。

引用

  1. Liang Y, Schott D, Zhang Y, et al. Auto-segmentation of pancreatic tumor in multi-parametric MRI using deep convolutional neural networks. Radiother Oncol. 2020. PMID: 32045787. doi:10.1016/j.radonc.2020.01.021

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。