Division of General Radiology, Department of Radiology, Medical University of Graz, Auenbruggerplatz 9, 8036, Graz, Austria.
Institute for Medical Informatics, Statistics and Documentation, Medical University of Graz, Auenbruggerplatz 2/9/V, 8036, Graz, Austria.
Eur Radiol. 2023 May;33(5):3064-3071. doi: 10.1007/s00330-023-09500-8. Epub 2023 Mar 22.
Reproducibility problems are a known limitation of radiomics. The segmentation of the target lesion plays a critical role in texture analysis variability. This study's aim was to compare the interobserver reliability of manual 2D vs. 3D lung lesion segmentation with and without pre-definition of the volume using a threshold of - 50 HU.
Seventy-five patients with histopathologically proven lung lesions (15 patients each with adenocarcinoma, squamous cell carcinoma, small cell lung cancer, carcinoid, and organizing pneumonia) who underwent an unenhanced CT scan of the chest were included. Three radiologists independently segmented each lesion manually in 3D and 2D with and without pre-segmentation volume definition by a HU threshold, and shape parameters and original, Laplacian of Gaussian-filtered, and wavelet-based texture features were derived. To assess interobserver reliability and identify the most robust texture features, intraclass correlation coefficients (ICCs) for different segmentation settings were calculated.
Shape parameters had high reliability (64-79% had excellent and good ICCs). Texture features had weak reliability levels, with the highest ICCs (38% excellent or good) found for original features in 3D segmentation without the use of a HU threshold. A small proportion (4.3-11.5%) of texture features had excellent or good ICC values at all segmentation settings.
Interobserver reliability of texture features from CT scans of a heterogeneous collection of manually segmented lung lesions was low with a small proportion of features demonstrating high reliability independent of the segmentation settings. These results indicate a limited applicability of texture analysis and the need to define robust texture features in patients with lung lesions.
• Our study showed a low reproducibility of texture features when 3 radiologists independently segmented lung lesions in CT images, which highlights a serious limitation of texture analysis. • Interobserver reliability of texture features was low regardless of whether the lesion was segmented in 2D and 3D with or without a HU threshold. • In contrast to texture features, shape parameters showed a high interobserver reliability when lesions were segmented in 2D vs. 3D with and without a HU threshold of - 50.
重测性问题是放射组学的一个已知局限性。目标病变的分割在纹理分析变异性中起着关键作用。本研究的目的是比较手动 2D 与 3D 肺病变分割的观察者间可靠性，以及有无使用 -50 HU 阈值预定义体积。
共纳入 75 例经组织病理学证实的肺病变患者（每个腺癌、鳞状细胞癌、小细胞肺癌、类癌和机化性肺炎各 15 例），这些患者均接受了胸部未增强 CT 扫描。3 名放射科医生分别独立地在 3D 和 2D 中手动分割每个病变，有无通过 HU 阈值预定义体积，以及提取形状参数和原始、拉普拉斯高斯滤波和小波基纹理特征。为了评估观察者间的可靠性并确定最稳健的纹理特征，计算了不同分割设置的组内相关系数（ICC）。
形状参数具有较高的可靠性（64%-79%的 ICC 为优秀或良好）。纹理特征的可靠性水平较低，在未使用 HU 阈值的 3D 分割中，原始特征的 ICC 最高（38%为优秀或良好）。在所有分割设置中，只有一小部分（4.3%-11.5%）的纹理特征具有优秀或良好的 ICC 值。
对一组异质的手动分割肺病变 CT 扫描进行分析，纹理特征的观察者间可靠性较低，只有一小部分特征在不依赖于分割设置的情况下具有较高的可靠性。这些结果表明纹理分析的适用性有限，需要为肺病变患者定义稳健的纹理特征。
• 我们的研究表明，当 3 名放射科医生在 CT 图像中独立分割肺病变时，纹理特征的可重复性较低，这突显了纹理分析的严重局限性。• 无论是否使用 HU 阈值在 2D 和 3D 中分割病变，纹理特征的观察者间可靠性均较低。• 与纹理特征相比，当在 2D 与 3D 中分割病变，有无使用 -50 HU 阈值时，形状参数的观察者间可靠性较高。
