文献速递泌尿生殖腹部MR

基于模型与深度学习联合的T2加权MRI前列腺三维分区自动分割:临床评估

Combined model-based and deep learning-based automated 3D zonal segmentation of the prostate on T2-weighted MR images: clinical evaluation.
PubMed 文献 · Eur Radiol · 2022-01-10

研究在NCI-ISBI-2013数据集和本中心70例T2加权数据上训练分区分割算法,在未参与训练的两台扫描仪(GE 1.5 T、Philips 3 T)的81例测试集上评估。算法全腺体、中央腺体和前纤维肌肉基质的中位Dice系数分别为96.4/97.4%、91.8/93.0%和79.9/89.6%,落在两名放射科医生间变异范围内,但受扫描仪和初始分割是否位于前列腺外显著影响。

要点

  • 算法全腺体、中央腺体和前纤维肌肉基质的中位Dice系数分别为96.4/97.4%、91.8/93.0%和79.9/89.6%,与两名医生间变异相当。
  • 扫描仪显著影响三个分割任务的平均Dice系数及其精度。
  • 13/81例初始分割始于前列腺外,经裁剪图像后仍可完成分区分割;此类病例平均Dice系数显著降低。

核心结论

  • 算法分割性能落在两名放射科医生间变异范围内,但仍受所用扫描仪显著影响;前纤维肌肉基质分割在不同患者间变异大,两名医生间变异也大。

研究设计与样本

  • 将已用于全腺体分割的模型与深度学习联合算法在NCI-ISBI-2013数据集和学术中心70例T2加权数据上训练分区分割;测试集从本中心两台未参与训练的扫描仪(GE 1.5 T、Philips 3 T)检查中随机选取82例、排除1例,由两名独立放射科医生校正自动分割,用beta回归评估影响平均Dice系数及其精度的因素。

诊断标准与关键数值

  • 算法中位Dice系数:全腺体96.4/97.4%、中央腺体91.8/93.0%、前纤维肌肉基质79.9/89.6%(两名参考医生)。
  • 两名放射科医生间Dice系数:全腺体95.8%、中央腺体93.6%、前纤维肌肉基质81.7%。
  • 13/81例初始分割始于前列腺外,经裁剪图像后完成分区分割,此类病例平均Dice系数显著降低。
  • 中央腺体分割中,较大前列腺的平均Dice系数显著降低。
  • 参考医生的选择除前纤维肌肉基质分割外无显著影响。

临床应用与阅片要点

  • 自动分区分割结果需由放射科医生校正,全腺体和中央腺体与医生间变异相当,前纤维肌肉基质变异较大。
  • 初始分割位于前列腺外时,裁剪图像后重新分割仍可完成分区分割,但平均Dice系数显著降低。
  • 不同扫描仪(GE 1.5 T与Philips 3 T)对算法性能有显著影响,阅片时需结合设备条件评估分割结果。

局限性

  • 算法性能虽落在医生间变异范围内,但仍受所用扫描仪显著影响。
  • 前纤维肌肉基质分割在不同患者间高度变异,两名放射科医生间也呈高变异。

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

OBJECTIVE: To train and to test for prostate zonal segmentation an existing algorithm already trained for whole-gland segmentation.

METHODS: The algorithm, combining model-based and deep learning-based approaches, was trained for zonal segmentation using the NCI-ISBI-2013 dataset and 70 T2-weighted datasets acquired at an academic centre. Test datasets were randomly selected among examinations performed at this centre on one of two scanners (General Electric, 1.5 T; Philips, 3 T) not used for training. Automated segmentations were corrected by two independent radiologists. When segmentation was initiated outside the prostate, images were cropped and segmentation repeated. Factors influencing the algorithm's mean Dice similarity coefficient (DSC) and its precision were assessed using beta regression.

RESULTS: Eighty-two test datasets were selected; one was excluded. In 13/81 datasets, segmentation started outside the prostate, but zonal segmentation was possible after image cropping. Depending on the radiologist chosen as reference, algorithm's median DSCs were 96.4/97.4%, 91.8/93.0% and 79.9/89.6% for whole-gland, central gland and anterior fibromuscular stroma (AFMS) segmentations, respectively. DSCs comparing radiologists' delineations were 95.8%, 93.6% and 81.7%, respectively. For all segmentation tasks, the scanner used for imaging significantly influenced the mean DSC and its precision, and the mean DSC was significantly lower in cases with initial segmentation outside the prostate. For central gland segmentation, the mean DSC was also significantly lower in larger prostates. The radiologist chosen as reference had no significant impact, except for AFMS segmentation.

CONCLUSIONS: The algorithm performance fell within the range of inter-reader variability but remained significantly impacted by the scanner used for imaging.

KEY POINTS: • Median Dice similarity coefficients obtained by the algorithm fell within human inter-reader variability for the three segmentation tasks (whole gland, central gland, anterior fibromuscular stroma). • The scanner used for imaging significantly impacted the performance of the automated segmentation for the three segmentation tasks. • The performance of the automated segmentation of the anterior fibromuscular stroma was highly variable across patients and showed also high variability across the two radiologists.

来源:Eur Radiol,查看原文。版权归原作者所有。

引用

  1. Rouvière, Moldovan, Vlachomitrou, et al. Combined model-based and deep learning-based automated 3D zonal segmentation of the prostate on T2-weighted MR images: clinical evaluation. Eur Radiol. 2022. PMID: 35001157. doi:10.1007/s00330-021-08408-5

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。