University Hospital Mannheim, Medical Faculty Mannheim, University of Heidelberg, Department of Radiation Oncology, Theodor Kutzer-Ufer 1-3, 68167, Mannheim, Germany; German Cancer Research Center (DKFZ), Division of Radiology, Im Neuenheimer Feld 280, 69120, Heidelberg, Germany.
German Cancer Research Center (DKFZ), Division of Medical Physics in Radiology, Im Neuenheimer Feld 280, 69120, Heidelberg, Germany.
Clin Radiol. 2020 Oct;75(10):798.e13-798.e22. doi: 10.1016/j.crad.2020.06.024. Epub 2020 Jul 25.
The aim of this study was to evaluate the robustness of radiomics features of a MRI (magnetic resonance imaging) phantom in quantitative diffusion-weighted imaging (DWI) and depending on the image resolution.
Scanning of an in-house developed DWI phantom was performed at a 1.5 T MRI scanner (Magnetom AERA, Siemens, Erlangen, Germany) using an echo planar imaging (EPI) DWI sequence (b=0,500,1,000 s/mm) with low (3×3 mm) and high (2×2 mm) image resolutions. Scans were repeated after phantom repositioning to evaluate retest reliability. Radiomics features were extracted after semi-automatic segmentation and standardised pre-processing. Intra-/interobserver reproducibility and test-retest robustness were assessed using intraclass correlation coefficients (ICC). Differences were tested with non-parametric Wilcoxon's signed-rank and Friedman's test (p < 0.05) with Dunn's post-hoc analysis.
Test-retest ICC was overall high with >0.90 for 39/46 radiomics features in all sequences/resolutions. Decreased test-retest ICCs were pronounced for conventional Min-value (overall ICC=0.817), and grey-level zone length matrix (GLZLM) features Short-Zone Emphasis (SZE) and Short-Zone Low Grey-level Emphasis (SZLGE) (for both overall ICC=0.927). Test-retest reproducibility was significantly different between b=500, 1,000 and apparent diffusion coefficient (ADC) (mean 0.975±0.050, 0.974±0.051 and 0.966±0.063), which remained significant after post-hoc analysis between b=1,000 and ADC (p = 0.022). ICCs were not significantly different between resolutions of 2×2 and 3×3 mm regarding b=500 (mean: 0.977±0.052 and 0.974±0.049, p = 0.612), b=1,000 (mean: 0.973±0.059 and 0.974±0.054, p = 0.516), and ADC (mean: 0.972±0.049 and 0.955±0.101, p = 0.851). Inter- and intra-observer reliability was consistently high for all sequences (overall mean 0.992±0.021 and 0.990±0.028).
Under ex-vivo conditions, DWI provided robust radiomics features with those from ADC being slightly less robust than from raw DWI (b=500, 1,000 s/mm). No significant difference was detected for different resolutions. Although, ex-vivo reliability of DWI radiomics features was high, no implications can be made regarding in-vivo analyses.
本研究旨在评估 MRI （磁共振成像）体模定量扩散加权成像（DWI）中放射组学特征的稳健性，并根据图像分辨率进行评估。
在 1.5 T MRI 扫描仪（Siemens，Erlangen，德国的 Magnetom AERA）上使用 EPI DWI 序列（b=0、500、1000 s/mm）对内部开发的 DWI 体模进行扫描，图像分辨率分别为低（3×3 mm）和高（2×2 mm）。在重新定位体模后进行重复扫描，以评估重测可靠性。在半自动分割和标准化预处理后提取放射组学特征。使用组内相关系数（ICC）评估内部/观察者重现性和测试-重测稳健性。使用非参数 Wilcoxon 符号秩检验和 Friedman 检验（p<0.05）以及 Dunn 事后分析测试差异。
所有序列/分辨率下，39/46 个放射组学特征的测试-重测 ICC 总体较高，均大于 0.90。常规 Min 值（整体 ICC=0.817）和灰度区长度矩阵（GLZLM）特征短区重点（SZE）和短区低灰度重点（SZLGE）的测试-重测 ICC 明显降低（两者整体 ICC=0.927）。b=500、1000 和表观扩散系数（ADC）之间的测试-重测再现性存在显著差异（平均值 0.975±0.050、0.974±0.051 和 0.966±0.063），在 b=1000 和 ADC 之间的事后分析中仍然显著（p=0.022）。b=500（平均值：0.977±0.052 和 0.974±0.049，p=0.612）、b=1000（平均值：0.973±0.059 和 0.974±0.054，p=0.516）和 ADC（平均值：0.972±0.049 和 0.955±0.101，p=0.851）时，分辨率为 2×2 和 3×3 mm 之间的 ICC 无显著差异。所有序列的观察者间和观察者内可靠性均较高（整体平均值 0.992±0.021 和 0.990±0.028）。
在离体条件下，DWI 提供了稳健的放射组学特征，ADC 略逊于原始 DWI（b=500、1000 s/mm）。不同分辨率之间没有发现显著差异。尽管 DWI 放射组学特征的离体可靠性较高，但不能对体内分析产生影响。
