Brown Kathryn H, Illyuk Jacob, Ghita Mihaela, Walls Gerard M, McGarry Conor K, Butterworth Karl T
Patrick G. Johnston Centre for Cancer Research, Queen's University Belfast, Belfast BT9 7AE, UK.
Northern Ireland Cancer Centre, Belfast Health & Social Care Trust, Belfast BT9 7JL, UK.
Cancers (Basel). 2023 May 9;15(10):2677. doi: 10.3390/cancers15102677.
Radiomics image analysis has the potential to uncover disease characteristics for the development of predictive signatures and personalised radiotherapy treatment. Inter-observer and inter-software delineation variabilities are known to have downstream effects on radiomics features, reducing the reliability of the analysis. The purpose of this study was to investigate the impact of these variabilities on radiomics outputs from preclinical cone-beam computed tomography (CBCT) scans. Inter-observer variabilities were assessed using manual and semi-automated contours of mouse lungs ( = 16). Inter-software variabilities were determined between two tools (3D Slicer and ITK-SNAP). The contours were compared using Dice similarity coefficient (DSC) scores and the 95th percentile of the Hausdorff distance (HD) metrics. The good reliability of the radiomics outputs was defined using intraclass correlation coefficients (ICC) and their 95% confidence intervals. The median DSC scores were high (0.82-0.94), and the HD metrics were within the submillimetre range for all comparisons. the shape and NGTDM features were impacted the most. Manual contours had the most reliable features (73%), followed by semi-automated (66%) and inter-software (51%) variabilities. From a total of 842 features, 314 robust features overlapped across all contouring methodologies. In addition, our results have a 70% overlap with features identified from clinical inter-observer studies.
放射组学图像分析有潜力揭示疾病特征，以开发预测性特征和个性化放射治疗方案。已知观察者间和软件间的轮廓勾画变异性会对放射组学特征产生下游影响，降低分析的可靠性。本研究的目的是调查这些变异性对临床前锥形束计算机断层扫描（CBCT）图像的放射组学输出的影响。使用小鼠肺部的手动和半自动轮廓（n = 16）评估观察者间变异性。在两种工具（3D Slicer和ITK-SNAP）之间确定软件间变异性。使用骰子相似系数（DSC）分数和豪斯多夫距离（HD）指标的第95百分位数比较轮廓。使用组内相关系数（ICC）及其95%置信区间定义放射组学输出的良好可靠性。所有比较的DSC分数中位数都很高（0.82 - 0.94），HD指标在亚毫米范围内。形状和邻域灰度差矩阵（NGTDM）特征受影响最大。手动轮廓具有最可靠的特征（73%），其次是半自动（66%）和软件间（51%）变异性。在总共842个特征中，314个稳健特征在所有轮廓勾画方法中重叠。此外，我们的结果与临床观察者间研究确定的特征有70%的重叠。
