文献速递腹部泌尿生殖MR

使用放射科医师评估的磁共振特征分类的机器学习模型预测常见实性肾肿瘤

Predicting common solid renal tumors using machine learning models of classification of radiologist-assessed magnetic resonance characteristics.
PubMed 文献 · Abdom Radiol (NY) · 2020-07-14

本研究回顾性评估330例实性肾肿块(SRM)的标准及多参数MR特征,用训练集建立递归分割(RP)、梯度提升机(GBM)和随机森林(RF)模型预测良恶性及亚型,并在测试集验证。GBM和RF预测良恶性的准确率为86%(95%CI 75%–93%),RF预测亚型准确率最高为81.2%(95%CI 69.5%–89.9%);各模型对透明细胞癌和乳头状肾细胞癌敏感度和特异度高,但对嫌色细胞癌、嗜酸细胞瘤和乏脂肪血管平滑肌脂肪瘤的鉴别能力有限。

要点

  • GBM和RF模型预测SRM良恶性的准确率达86%,RP模型为83%。
  • RF模型预测SRM亚型的准确率最高,为81.2%。
  • 各模型对嫌色细胞癌、嗜酸细胞瘤和乏脂肪血管平滑肌脂肪瘤的鉴别能力有限。

核心结论

  • 基于放射科医师评估的影像特征,集成方法预测SRM具有较高的良恶性鉴别准确率。
  • SRM亚型分类受限于对嫌色细胞癌、嗜酸细胞瘤和乏脂肪血管平滑肌脂肪瘤的区分能力。

研究设计与样本

  • 回顾性评估330例实性肾肿块的MR图像,分析标准及多参数特征。
  • 使用训练数据集开发预测良恶性及亚型的模型,并在测试数据集评估性能。
  • 采用递归分割(RP)、梯度提升机(GBM)和随机森林(RF)三种集成方法。

诊断标准与关键数值

  • 测试集中,GBM和RF模型预测良恶性SRM的准确率为86%(95%CI 75%–93%),RP模型为83%(95%CI 71%–91%)。
  • RF预测SRM亚型的准确率为81.2%(95%CI 69.5%–89.9%),GBM为73.4%(95%CI 60.9%–83.7%),RP为70.3%(95%CI 57.6%–81.1%)。
  • RF模型较RP模型边际同质性降低(P < 0.001),但与GBM模型无显著差异(P = 0.135)。
  • 所有模型对透明细胞癌和乳头状肾细胞癌均有高敏感度和特异度。

局限性

  • SRM亚型分类受限于对嫌色细胞癌、嗜酸细胞瘤和乏脂肪血管平滑肌脂肪瘤的区分能力。

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

PURPOSE: Solid renal masses (SRM) are difficult to differentiate based on standard MR features. The purpose of this study was to assess MR imaging features of SRM to evaluate performance of ensemble methods of classifying SRM subtypes.

MATERIALS AND METHODS: MR images of SRM (n = 330) were retrospectively evaluated for standard and multiparametric (mp) features. Models of MR features for predicting malignant and benign lesions as well as subtyping SRM were developed using a training dataset and performance was evaluated in a test data-set using recursive partitioning (RP), gradient booting machine (GBM), and random forest (RF) methods.

RESULTS: In the test dataset, GBM and RF models demonstrated an accuracy of 86% (95% CI 75% to 93%) for predicting benign versus malignant SRM compared to 83% (95% CI 71% to 91%) for the RP model. RF had the greatest accuracy in predicting SRM subtypes, 81.2% (95% CI 69.5% to 89.9%) compared with GBM 73.4% (95% CI 60.9% to 83.7%) or RP 70.3% (95% CI 57.6% to 81.1%). Marginal homogeneity was reduced by the RF model compared with the RP model (P < 0.001), but not the GBM model (P = 0.135). All models had high sensitivity and specificity for clear cell and papillary renal cell carcinomas (RCC), but performed less well in differentiating chromophobe RCC, oncocytomas, and fat-poor angiomyolipomas.

CONCLUSION: Ensemble methods for prediction of SRM from radiologist-assessed image characteristics have high accuracy for distinguishing benign and malignant lesions. SRM subtype classification is limited by the ability to categorize chromophobe RCCs, oncocytomas, and fat-poor angiomyolipomas.

来源:Abdom Radiol (NY),查看原文。版权归原作者所有。

引用

  1. Lopes Vendrami C, McCarthy RJ, Villavicencio CP, et al. Predicting common solid renal tumors using machine learning models of classification of radiologist-assessed magnetic resonance characteristics. Abdom Radiol (NY). 2020. PMID: 32666233. doi:10.1007/s00261-020-02637-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。