文献速递泌尿生殖腹部CT

基于CT影像组学模型术前预测透明细胞肾细胞癌肾纤维包膜侵犯

Preoperative prediction of renal fibrous capsule invasion in clear cell renal cell carcinoma using CT-based radiomics model.
PubMed 文献 · Br J Radiol · 2024-01-01

本研究回顾性纳入163例透明细胞肾细胞癌患者,利用皮髓质期CT图像半自动分割肾脏和肿瘤,比较10种机器学习分类器术前预测肾纤维包膜侵犯的性能,AdaBoost在批次1和批次2的AUC分别为0.83(95%CI 0.68-0.98)和0.74(95%CI 0.51-0.97),并发现9个共性显著特征。

要点

  • 基于皮髓质期CT影像组学的AdaBoost分类器术前预测肾纤维包膜侵犯批次1的AUC为0.83(95%CI 0.68-0.98)。
  • 肾脏分割Dice相似系数训练集0.98、测试集0.96,肿瘤分割训练集0.94、测试集0.86。
  • 从2个独立批次数据集中发现9个共性显著分类特征,包括形态学和纹理特征。

核心结论

  • 基于CT的影像组学分类器对透明细胞肾细胞癌纤维包膜侵犯的术前预测表现良好。

研究设计与样本

  • 回顾性研究,纳入163例在本院接受术前腹部增强CT和肾切除术的透明细胞肾细胞癌患者。

诊断标准与关键数值

  • 采用迁移学习从肾脏肿瘤分割挑战数据集获得基础模型,对皮髓质期CT图像半自动分割肾脏和肿瘤;比较10种机器学习分类器,以准确率、精确率、召回率和AUC评估性能;AdaBoost在批次1的准确率0.8571、精确率0.8333、召回率0.9091、F1分数0.8696,批次1和批次2的AUC分别为0.83(95%CI 0.68-0.98)和0.74(95%CI 0.51-0.97)。

临床应用与阅片要点

  • 作者指出腹部CT术前无创预测肾细胞癌肾纤维包膜侵犯相当困难,整合影像组学特征的机器学习分类器显示出辅助肾细胞癌患者手术方案选择的潜力。

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)。

英文原文(摘要)· 点击展开

OBJECTIVES: To develop radiomics-based classifiers for preoperative prediction of fibrous capsule invasion in renal cell carcinoma (RCC) patients by CT images.

METHODS: In this study, clear cell RCC (ccRCC) patients who underwent both preoperative abdominal contrast-enhanced CT and nephrectomy surgery at our hospital were analysed. By transfer learning, we used base model obtained from Kidney Tumour Segmentation challenge dataset to semi-automatically segment kidney and tumours from corticomedullary phase (CMP) CT images. Dice similarity coefficient (DSC) was measured to evaluate the performance of segmentation models. Ten machine learning classifiers were compared in our study. Performance of the models was assessed by their accuracy, precision, recall, and area under the receiver operating characteristic curve (AUC). The reporting and methodological quality of our study was assessed by the CLEAR checklist and METRICS score.

RESULTS: This retrospective study enrolled 163 ccRCC patients. The semiautomatic segmentation model using CMP CT images obtained DSCs of 0.98 in the training cohort and 0.96 in the test cohort for kidney segmentation, and DSCs of 0.94 and 0.86 for tumour segmentation in the training and test set, respectively. For preoperative prediction of renal capsule invasion, the AdaBoost had the best performance in batch 1, with accuracy, precision, recall, and F1-score equal to 0.8571, 0.8333, 0.9091, and 0.8696, respectively; and the same classifier was also the most suitable for this classification in batch 2. The AUCs of AdaBoost for batch 1 and batch 2 were 0.83 (95% CI: 0.68-0.98) and 0.74 (95% CI: 0.51-0.97), respectively. Nine common significant features for classification were found from 2 independent batch datasets, including morphological and texture features.

CONCLUSIONS: The CT-based radiomics classifiers performed well for the preoperative prediction of fibrous capsule invasion in ccRCC.

ADVANCES IN KNOWLEDGE: Noninvasive prediction of renal fibrous capsule invasion in RCC is rather difficult by abdominal CT images before surgery. A machine learning classifier integrated with radiomics features shows a promising potential to assist surgical treatment options for RCC patients.

来源:Br J Radiol,查看原文。版权归原作者所有。

引用

  1. Zhang Y, Zhao J, Li Z, et al. Preoperative prediction of renal fibrous capsule invasion in clear cell renal cell carcinoma using CT-based radiomics model. Br J Radiol. 2024. PMID: 38897659. doi:10.1093/bjr/tqae122

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。