文献速递泌尿生殖与妇科肿瘤放疗放疗物理与技术

多中心数据集对基于卷积神经网络的18F-PSMA-1007 PET原发前列腺癌自动肿瘤勾画的影响

The impact of multicentric datasets for the automated tumor delineation in primary prostate cancer using convolutional neural networks on 18F-PSMA-1007 PET.
PubMed 文献 · Radiat Oncol · 2024-08-07

本研究用161例来自四家机构的18F-PSMA-1007 PET图像训练nnU-Net,采用五折交叉验证比较单中心与多中心训练对前列腺内GTV自动勾画的影响。混合多中心训练中位DSC为0.76,与逐中心排除训练的中位DSC 0.74相比无显著差异(p=0.18);Dresden和Cyprus队列在多中心训练下DSC显著提高。

要点

  • 多中心训练中位DSC 0.76(IQR 0.64-0.84) vs 逐中心排除训练0.74(IQR 0.56-0.86),p=0.18。
  • Dresden队列多中心训练DSC 0.71 vs 单中心0.68,p<0.001。
  • Cyprus队列多中心训练DSC 0.74 vs 单中心0.72,p<0.01。

核心结论

  • 基于多中心多样数据集训练的CNN对前列腺内18F-PSMA-1007 PET GTV自动勾画在多数情况下能泛化至其他中心未见数据。
  • 与仅用单中心数据集训练相比,多中心数据集训练可提高前列腺内18F-PSMA-1007 PET GTV分割性能。
  • 同一CNN的分割性能可因训练和测试所用数据集不同而变化。

研究设计与样本

  • 采用nnU-Net,使用来自四家机构的161例18F-PSMA-1007 PET图像进行训练与测试(Freiburg n=96、Munich n=19、Cyprus n=32、Dresden n=14)。
  • 数据集划分时系统性地将每个中心的数据排除出训练并仅用于测试,以评估模型对不熟悉来源数据的泛化性与适应性。
  • 通过五折交叉验证比较单中心训练模型与汇总多中心训练模型的性能,主要评价指标为Dice相似系数、Hausdorff距离和体积分析。

主要发现

  • 混合训练方法在五折交叉验证中中位DSC为0.76(IQR 0.64-0.84),与逐中心排除数据训练的模型中位DSC 0.74(IQR 0.56-0.86)相比无显著差异(p=0.18)。
  • Dresden队列多中心训练中位DSC 0.71(IQR 0.58-0.80) vs 单中心训练0.68(IQR 0.50-0.80),p<0.001。
  • Cyprus队列多中心训练中位DSC 0.74(IQR 0.62-0.83) vs 单中心训练0.72(IQR 0.54-0.82),p<0.01。
  • Munich队列多中心训练DSC 0.74(IQR 0.60-0.80) vs 单中心训练0.72(IQR 0.59-0.82),p>0.05,未达统计学显著。
  • Freiburg队列多中心训练DSC 0.78(IQR 0.53-0.87) vs 单中心训练0.71(IQR 0.53-0.83),p=0.23,未达统计学显著。

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: Convolutional Neural Networks (CNNs) have emerged as transformative tools in the field of radiation oncology, significantly advancing the precision of contouring practices. However, the adaptability of these algorithms across diverse scanners, institutions, and imaging protocols remains a considerable obstacle. This study aims to investigate the effects of incorporating institution-specific datasets into the training regimen of CNNs to assess their generalization ability in real-world clinical environments. Focusing on a data-centric analysis, the influence of varying multi- and single center training approaches on algorithm performance is conducted. METHODS: nnU-Net is trained using a dataset comprising 161 18F-PSMA-1007 PET images collected from four distinct institutions (Freiburg: n = 96, Munich: n = 19, Cyprus: n = 32, Dresden: n = 14). The dataset is partitioned such that data from each center are systematically excluded from training and used solely for testing to assess the model's generalizability and adaptability to data from unfamiliar sources. Performance is compared through a 5-Fold Cross-Validation, providing a detailed comparison between models trained on datasets from single centers to those trained on aggregated multi-center datasets. Dice Similarity Score, Hausdorff distance and volumetric analysis are used as primary evaluation metrics. RESULTS: The mixed training approach yielded a median DSC of 0.76 (IQR: 0.64-0.84) in a five-fold cross-validation, showing no significant differences (p = 0.18) compared to models trained with data exclusion from each center, which performed with a median DSC of 0.74 (IQR: 0.56-0.86). Significant performance improvements regarding multi-center training were observed for the Dresden cohort (multi-center median DSC 0.71, IQR: 0.58-0.80 vs. single-center 0.68, IQR: 0.50-0.80, p < 0.001) and Cyprus cohort (multi-center 0.74, IQR: 0.62-0.83 vs. single-center 0.72, IQR: 0.54-0.82, p < 0.01). While Munich and Freiburg also showed performance improvements with multi-center training, results showed no statistical significance (Munich: multi-center DSC 0.74, IQR: 0.60-0.80 vs. single-center 0.72, IQR: 0.59-0.82, p > 0.05; Freiburg: multi-center 0.78, IQR: 0.53-0.87 vs. single-center 0.71, IQR: 0.53-0.83, p = 0.23). CONCLUSION: CNNs trained for auto contouring intraprostatic GTV in 18F-PSMA-1007 PET on a diverse dataset from multiple centers mostly generalize well to unseen data from other centers. Training on a multicentric dataset can improve performance compared to training exclusively with a single-center dataset regarding intraprostatic 18F-PSMA-1007 PET GTV segmentation. The segmentation performance of the same CNN can vary depending on the dataset employed for training and testing.

来源:Radiat Oncol,查看原文。版权归原作者所有。

引用

  1. Holzschuh JC, Mix M, Freitag MT, et al. The impact of multicentric datasets for the automated tumor delineation in primary prostate cancer using convolutional neural networks on 18F-PSMA-1007 PET. Radiat Oncol. 2024. PMID: 39113123. doi:10.1186/s13014-024-02491-w

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。