Robustness versus disease differentiation when varying parameter settings in radiomics features: application to nasopharyngeal PET/CT
Robustness versus disease differentiation when varying parameter settings in radiomics features: application to nasopharyngeal PET/CT
复制标题
改变放射组学特征参数设置时的鲁棒性与疾病分化：在鼻咽 PET/CT 中的应用
DOI：
10.1007/s00330-018-5343-0
复制
发表时间：
2018-08-01
影响因子：
5.9
通讯作者：
Lu, Lijun
中科院分区：
文献类型：
--
作者：
Lv, Wenbing;Yuan, Qingyu;Lu, Lijun
ObjectivesTo调查参数设置的影响，用于产生放射组学功能的鲁棒性和疾病分化（鼻咽癌（NPC）与慢性鼻咽炎（CN）在FDG PET/CT imaging）。MethodsWe研究了106例患者（69/37 NPC/CN，病理证实），并提取57放射组学功能在不同的参数设置。通过类内相关系数（ICC）评估耐用性。逻辑回归与留一交叉验证被用来产生分类概率，和诊断性能进行了评估的接收器工作特征曲线（AUC）下的面积。ResultsVariing平均策略和对称性，4/26 GLCM功能显示不良的成对ICC范围为0.02-0.98，而描绘良好的AUC为0.82-0.91。不同的距离，5/26个GLCM特征显示ICC为0.82-0.99，而相应的AUC为0.52-0.91。6/13的GLRLM特征显示关于平均策略的高AUC（0.81-0.89）和高ICC（0.85-0.99）。7/13的GLSZM特征显示AUC为0.81-0.90，而在不同邻域下的ICC为0.01-0.99。2/5的NGTDM特征显示AUC为0.81-0.85，而不同窗口大小的ICC为0.19-0.89。区分一个子集的NPC（阶段I-II）CN，两个SumEntropy和SZLGE实现了显着更高的AUC比代谢活跃的肿瘤体积（AUC：0.91 vs.0.72，p <0.01）。因此，不应过分强调放射组学特征的鲁棒性，以消除对临床任务评估的特征。在区分鼻咽癌和CN时，一些放射组学特征（例如SumEntropy、SZLGE、LGZE）优于传统指标。关键点·稳健性差不一定意味着区分性能差。·不应过分强调放射组学特征的绝对尺度稳健性。·Radiomics的SumEntropy，SZLGE和LGZE优于传统指标。
ObjectivesTo investigate the impact of parameter settings as used for the generation of radiomics features on their robustness and disease differentiation (nasopharyngeal carcinoma (NPC) versus chronic nasopharyngitis (CN) in FDG PET/CT imaging).MethodsWe studied 106 patients (69/37 NPC/CN, pathology confirmed), and extracted 57 radiomics features under different parameter settings. Robustness was assessed by the intra-class correlation coefficient (ICC). Logistic regression with leave-one-out cross validation was used to generate classification probabilities, and diagnostic performance was assessed by the area under the receiver operating characteristic curve (AUC).ResultsVarying averaging strategies and symmetry, 4/26 GLCM features showed poor range of pairwise ICCs of 0.02–0.98, while depicting good AUCs of 0.82–0.91. Varying distances, 5/26 GLCM features showed ICCs of 0.82–0.99 while corresponding AUCs were 0.52–0.91. 6/13 GLRLM features showed both high AUC (0.81–0.89) and high ICC (0.85–0.99) regarding to averaging strategies. 7/13 GLSZM features showed AUCs of 0.81–0.90 while having ICCs of 0.01–0.99 under different neighbourhoods. 2/5 NGTDM features showed AUCs of 0.81–0.85 while having ICCs of 0.19–0.89 for different window sizes. Differentiating a subset of NPC (stages I–II) form CN, both SumEntropy and SZLGE achieved significantly higher AUCs than metabolically active tumour volume (AUC: 0.91 vs. 0.72,p<0.01).ConclusionsRadiomics features depicting poor absolute-scale robustness regarding to parameter settings can still lead to good diagnostic performance. As such, robustness of radiomics features should not be overemphasized for removal of features towards assessment of clinical tasks. For differentiating NPC from CN, some radiomics features (e.g. SumEntropy, SZLGE, LGZE) outperformed conventional metrics.Key Points• Poor robustness did not necessarily translate into poor differentiation performance.• Absolute-scale robustness of radiomics features should not be overemphasized.• Radiomics features SumEntropy, SZLGE and LGZE outperformed conventional metrics.