文献速递肿瘤 PET/CT

基于18F-FDG-PET/CT的深度学习模型用于胰腺导管腺癌术前病理分级的全自动预测

18F-FDG-PET/CT-based deep learning model for fully automated prediction of pathological grading for pancreatic ductal adenocarcinoma before surgery.
PubMed 文献 · EJNMMI Res · 2023-05-25

本研究回顾性纳入2016年1月至2021年9月的370例胰腺导管腺癌患者,基于术前18F-FDG-PET/CT开发深度学习模型全自动预测病理分级。肿瘤分割模型Dice评分为0.89;基于分割特征的预测模型AUC为0.74,整合关键临床数据后AUC提升至0.77。作者称这是首个以内到端全自动方式预测胰腺导管腺癌病理分级的深度学习模型。

要点

  • 基于PET/CT的肿瘤分割模型对胰腺导管腺癌的Dice评分为0.89。
  • 基于分割特征的预测模型AUC为0.74,准确度0.72、灵敏度0.73、特异度0.72。
  • 整合关键临床数据后AUC提升至0.77,准确度0.75、灵敏度0.77、特异度0.73。

核心结论

  • 作者称这是首个以内到端全自动方式预测胰腺导管腺癌病理分级的深度学习模型,有望改善临床决策。

研究设计与样本

  • 回顾性纳入2016年1月至2021年9月的370例胰腺导管腺癌患者,所有患者术前接受18F-FDG-PET/CT检查、术后获得病理结果。
  • 先用其中100例开发胰腺癌病灶分割深度学习模型并应用于其余病例以获得病灶区域。
  • 将所有患者按5:1:1比例分为训练集、验证集和测试集,利用病灶区域计算的特征和患者关键临床特征开发病理分级预测模型。
  • 通过七折交叉验证验证模型稳定性。

主要发现

  • 开发的基于PET/CT的胰腺导管腺癌肿瘤分割模型Dice评分为0.89。
  • 基于分割模型开发的PET/CT深度学习模型AUC为0.74,准确度0.72、灵敏度0.73、特异度0.72。
  • 整合关键临床数据后模型AUC提升至0.77,准确度、灵敏度、特异度分别提升至0.75、0.77、0.73。

关键数值与推荐

  • 肿瘤分割模型Dice评分0.89。
  • 基于分割特征的预测模型AUC 0.74,准确度0.72、灵敏度0.73、特异度0.72。
  • 整合临床数据后模型AUC 0.77,准确度0.75、灵敏度0.77、特异度0.73。

临床怎么用

  • 原文未据此给出推荐。

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)。

原文

BACKGROUND: The determination of pathological grading has a guiding significance for the treatment of pancreatic ductal adenocarcinoma (PDAC) patients. However, there is a lack of an accurate and safe method to obtain pathological grading before surgery. The aim of this study is to develop a deep learning (DL) model based on 18F-fluorodeoxyglucose-positron emission tomography/computed tomography (18F-FDG-PET/CT) for a fully automatic prediction of preoperative pathological grading of pancreatic cancer. METHODS: A total of 370 PDAC patients from January 2016 to September 2021 were collected retrospectively. All patients underwent 18F-FDG-PET/CT examination before surgery and obtained pathological results after surgery. A DL model for pancreatic cancer lesion segmentation was first developed using 100 of these cases and applied to the remaining cases to obtain lesion regions. After that, all patients were divided into training set, validation set, and test set according to the ratio of 5:1:1. A predictive model of pancreatic cancer pathological grade was developed using the features computed from the lesion regions obtained by the lesion segmentation model and key clinical characteristics of the patients. Finally, the stability of the model was verified by sevenfold cross-validation. RESULTS: The Dice score of the developed PET/CT-based tumor segmentation model for PDAC was 0.89. The area under curve (AUC) of the PET/CT-based DL model developed on the basis of the segmentation model was 0.74, with an accuracy, sensitivity, and specificity of 0.72, 0.73, and 0.72, respectively. After integrating key clinical data, the AUC of the model improved to 0.77, with its accuracy, sensitivity, and specificity boosted to 0.75, 0.77, and 0.73, respectively. CONCLUSION: To the best of our knowledge, this is the first deep learning model to end-to-end predict the pathological grading of PDAC in a fully automatic manner, which is expected to improve clinical decision-making.

来源:EJNMMI Res,查看原文。版权归原作者所有。

引用

  1. Zhang G, Bao C, Liu Y, et al. 18F-FDG-PET/CT-based deep learning model for fully automated prediction of pathological grading for pancreatic ductal adenocarcinoma before surgery. EJNMMI Res. 2023. PMID: 37231321. doi:10.1186/s13550-023-00985-4

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。