School of Data Science, North University of China, 3 Xueyuan Road, Taiyuan, Shanxi 030051, China.
Shanxi Province Cancer Hospital, 3 Zhigong New Street, Taiyuan, Shanxi 030013, China.
Biomed Res Int. 2021 Nov 15;2021:5522452. doi: 10.1155/2021/5522452. eCollection 2021.
To evaluate the utility of radiomics features in differentiating central lung cancers and atelectasis on contrast-enhanced computed tomography (CT) images. This study is retrospective.
In this study, 36 patients with central pulmonary cancer and atelectasis between July 2013 and June 2018 were identified. A total of 1,653 2D and 2,327 3D radiomics features were extracted from segmented lung cancers and atelectasis on contrast-enhanced CT. The refined features were investigated for usefulness in classifying lung cancer and atelectasis according to the information gain, and 10 models were trained based on these features. The classification model is trained and tested at the region level and pixel level, respectively.
Among all the extracted features, 334 2D features and 1,507 3D features had an information gain (IG) greater than 0.1. The highest accuracy (AC) of the region classifiers was 0.9375. The best Dice score, Hausdorff distance, and voxel AC were 0.2076, 45.28, and 0.8675, respectively.
Radiomics features derived from contrast-enhanced CT images can differentiate lung cancers and atelectasis at the regional and voxel levels.
评估 CT 增强图像中放射组学特征在区分中央型肺癌和肺不张中的应用价值。本研究为回顾性研究。
本研究纳入了 2013 年 7 月至 2018 年 6 月期间的 36 例中央型肺癌和肺不张患者。从增强 CT 图像中分割出的肺癌和肺不张区域提取了 1653 个 2D 放射组学特征和 2327 个 3D 放射组学特征。根据信息增益对这些特征进行了筛选，以评估其在区分肺癌和肺不张中的作用，并基于这些特征训练了 10 个模型。分类模型分别在区域水平和像素水平进行训练和测试。
在所提取的特征中，有 334 个 2D 特征和 1507 个 3D 特征的信息增益（IG）大于 0.1。区域分类器的最高准确率（AC）为 0.9375。最佳 Dice 评分、Hausdorff 距离和体素 AC 分别为 0.2076、45.28 和 0.8675。
从 CT 增强图像中提取的放射组学特征可在区域和体素水平上区分肺癌和肺不张。
