文献速递肿瘤 PET/CT

基于术前[18F]FDG PET/CT多机器学习模型预测病理性高侵袭性肺癌

Predicting pathological highly invasive lung cancer from preoperative [18F]FDG PET/CT with multiple machine learning models.
PubMed 文献 · Eur J Nucl Med Mol Imaging · 2022-11-17

研究纳入873例因原发性肺癌行肺叶或肺段切除术的患者,从术前PET/CT提取影像组学特征构建7种机器学习模型及集成模型(ENS),经100次迭代评估。ENS在测试集平均AUC为0.880,嵌套交叉验证AUC为0.882(95%CI 0.860-0.905),决策曲线分析阈值概率3%至98%时有效,准确率较CTR提升超过8%。

要点

  • 基于术前PET/CT影像组学的ENS在测试集平均AUC为0.880,标准差0.0165
  • 嵌套交叉验证AUC为0.882(95%CI 0.860-0.905),Brier评分0.131
  • ENS准确率较CTR提升超过8%,决策曲线阈值概率3%至98%时有效

核心结论

  • 基于术前[18F]FDG PET/CT图像的机器学习模型能高判别力且稳定地预测病理性高侵袭性肺癌。
  • 校准图显示良好一致性,提示该模型可用于定量风险评估。

研究设计与样本

  • 共纳入873例因原发性肺癌行肺叶或肺段切除术的患者。
  • 使用PyRadiomics包从术前PET/CT图像提取影像组学特征。
  • 评估7种机器学习模型及全部模型的集成(ENS),经100次迭代。
  • 在嵌套交叉验证中计算高侵袭性肺癌概率,评估校准图和临床实用性,并与CT图像上的实性成分肿瘤比(CTR)比较。

主要发现

  • 训练集中,PET与CT特征联合时所有模型AUC均≥0.880。
  • 测试集中ENS平均AUC最高为0.880,标准差最小为0.0165; cutoff为0.5时准确率0.804、F1 0.851、精确率0.821、召回率0.885。
  • 嵌套交叉验证中AUC为0.882(95%CI 0.860-0.905),校准图显示一致性,Brier评分0.131。
  • 决策曲线分析显示ENS在阈值概率3%至98%范围内有效。
  • ENS准确率较CTR提升超过8%。

关键数值与推荐

  • 测试集ENS平均AUC 0.880,标准差0.0165。
  • 嵌套交叉验证AUC 0.882(95%CI 0.860-0.905),Brier评分0.131。
  • ENS cutoff为0.5时准确率0.804、F1 0.851、精确率0.821、召回率0.885。
  • ENS决策曲线有效阈值概率范围3%至98%。
  • ENS准确率较CTR提升超过8%。

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

PURPOSE: The efficacy of sublobar resection of primary lung cancer have been proven in recent years. However, sublobar resection for highly invasive lung cancer increases local recurrence. We developed and validated multiple machine learning models predicting pathological invasiveness of lung cancer based on preoperative [18F]fluorodeoxyglucose (FDG) positron emission tomography (PET) and computed tomography (CT) radiomic features. METHODS: Overall, 873 patients who underwent lobectomy or segmentectomy for primary lung cancer were enrolled. Radiomics features were extracted from preoperative PET/CT images with the PyRadiomics package. Seven machine learning models and an ensemble of all models (ENS) were evaluated after 100 iterations. In addition, the probability of highly invasive lung cancer was calculated in a nested cross-validation to assess the calibration plot and clinical usefulness and to compare to consolidation tumour ratio (CTR) on CT images, one of the generally used diagnostic criteria. RESULTS: In the training set, when PET and CT features were combined, all models achieved an area under the curve (AUC) of ≥ 0.880. In the test set, ENS showed the highest mean AUC of 0.880 and smallest standard deviation of 0.0165, and when the cutoff was 0.5, accuracy of 0.804, F1 of 0.851, precision of 0.821, and recall of 0.885. In the nested cross-validation, the AUC of 0.882 (95% CI: 0.860-0.905) showed a high discriminative ability, and the calibration plot indicated consistency with a Brier score of 0.131. A decision curve analysis showed that the ENS was valid with a threshold probability ranging from 3 to 98%. Accuracy showed an improvement of more than 8% over the CTR. CONCLUSION: The machine learning model based on preoperative [18F]FDG PET/CT images was able to predict pathological highly invasive lung cancer with high discriminative ability and stability. The calibration plot showed good consistency, suggesting its usefulness in quantitative risk assessment.

来源:Eur J Nucl Med Mol Imaging,查看原文。版权归原作者所有。

引用

  1. Onozato Y, Iwata T, Uematsu Y, et al. Predicting pathological highly invasive lung cancer from preoperative [18F]FDG PET/CT with multiple machine learning models. Eur J Nucl Med Mol Imaging. 2022. PMID: 36385219. doi:10.1007/s00259-022-06038-7

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。