Department of Radiology, Research Institute of Radiological Science, Center for Clinical Imaging Data Science, Severance Hospital, Yonsei University College of Medicine, Seoul, Korea.
Korean J Radiol. 2021 Mar;22(3):334-343. doi: 10.3348/kjr.2020.0099. Epub 2020 Nov 3.
We aimed to develop a prediction model for diagnosing severe aortic stenosis (AS) using computed tomography (CT) radiomics features of aortic valve calcium (AVC) and machine learning (ML) algorithms.
We retrospectively enrolled 408 patients who underwent cardiac CT between March 2010 and August 2017 and had echocardiographic examinations (240 patients with severe AS on echocardiography [the severe AS group] and 168 patients without severe AS [the non-severe AS group]). Data were divided into a training set (312 patients) and a validation set (96 patients). Using non-contrast-enhanced cardiac CT scans, AVC was segmented, and 128 radiomics features for AVC were extracted. After feature selection was performed with three ML algorithms (least absolute shrinkage and selection operator [LASSO], random forests [RFs], and eXtreme Gradient Boosting [XGBoost]), model classifiers for diagnosing severe AS on echocardiography were developed in combination with three different model classifier methods (logistic regression, RF, and XGBoost). The performance (c-index) of each radiomics prediction model was compared with predictions based on AVC volume and score.
The radiomics scores derived from LASSO were significantly different between the severe AS and non-severe AS groups in the validation set (median, 1.563 vs. 0.197, respectively, < 0.001). A radiomics prediction model based on feature selection by LASSO + model classifier by XGBoost showed the highest c-index of 0.921 (95% confidence interval [CI], 0.869-0.973) in the validation set. Compared to prediction models based on AVC volume and score (c-indexes of 0.894 [95% CI, 0.815-0.948] and 0.899 [95% CI, 0.820-0.951], respectively), eight and three of the nine radiomics prediction models showed higher discrimination abilities for severe AS. However, the differences were not statistically significant ( > 0.05 for all).
Models based on the radiomics features of AVC and ML algorithms may perform well for diagnosing severe AS, but the added value compared to AVC volume and score should be investigated further.
我们旨在使用主动脉瓣钙（AVC）的计算机断层扫描（CT）放射组学特征和机器学习（ML）算法开发一种用于诊断严重主动脉瓣狭窄（AS）的预测模型。
我们回顾性地招募了 2010 年 3 月至 2017 年 8 月期间接受心脏 CT 检查且经超声心动图检查的 408 例患者（超声心动图上 240 例为严重 AS [严重 AS 组]，168 例为非严重 AS [非严重 AS 组]）。数据分为训练集（312 例）和验证集（96 例）。使用非增强型心脏 CT 扫描对 AVC 进行分割，并提取 128 个 AVC 放射组学特征。使用三种 ML 算法（最小绝对收缩和选择算子[LASSO]、随机森林[RFs]和极端梯度提升[XGBoost]）进行特征选择后，结合三种不同的模型分类器方法（逻辑回归、RF 和 XGBoost）开发用于诊断超声心动图上严重 AS 的模型分类器。比较每个放射组学预测模型的性能（C 指数）与基于 AVC 体积和评分的预测结果。
验证集中，严重 AS 组和非严重 AS 组的 LASSO 得出的放射组学评分差异有统计学意义（中位数分别为 1.563 与 0.197， < 0.001）。基于 LASSO 特征选择+XGBoost 模型分类器的放射组学预测模型在验证集中的 C 指数最高，为 0.921（95%置信区间[CI]，0.869-0.973）。与基于 AVC 体积和评分的预测模型（C 指数分别为 0.894[95%CI，0.815-0.948]和 0.899[95%CI，0.820-0.951]）相比，9 个放射组学预测模型中有 8 个和 3 个对严重 AS 的鉴别能力更高。然而，差异无统计学意义（均 > 0.05）。
基于 AVC 放射组学特征和 ML 算法的模型可能对诊断严重 AS 具有良好的效果，但与 AVC 体积和评分相比，其增值作用尚需进一步研究。
