Cancer radiomics is an emerging discipline promising to elucidate lesion phenotypes and tumour heterogeneity through patterns of enhancement, texture, morphology and shape. The prevailing technique for image texture analysis relies on the construction and synthesis of grey‐level co‐occurrence matrices (GLCM). Practice currently reduces the structured count data of a GLCM to reductive and redundant summary statistics for which analysis requires variable selection and multiple comparisons for each application, thus limiting reproducibility. In this article, we develop a Bayesian multivariate probabilistic framework for the analysis and unsupervised clustering of a sample of GLCM objects. By appropriately accounting for skewness and zero inflation of the observed counts and simultaneously adjusting for existing spatial autocorrelation at nearby cells, the methodology facilitates estimation of texture pattern distributions within the GLCM lattice itself. The techniques are applied to cluster images of adrenal lesions obtained from CT scans with and without administration of contrast. We further assess whether the resultant subtypes are clinically oriented by investigating their correspondence with pathological diagnoses. Additionally, we compare performance to a class of machine learning approaches currently used in cancer radiomics with simulation studies.

中文翻译：

---

纹理模式异质性的贝叶斯非参数模型

癌症放射学是一门新兴学科，有望通过增强，纹理，形态和形状的模式阐明病灶表型和肿瘤异质性。图像纹理分析的主要技术依赖于灰度共现矩阵（GLCM）的构建和合成。当前，实践将GLCM的结构化计数数据缩减为归约和冗余的摘要统计信息，对于这些统计信息，分析需要针对每个应用程序进行变量选择和多次比较，从而限制了可重复性。在本文中，我们开发了贝叶斯多元概率框架，用于分析和无监督聚类GLCM对象。通过适当考虑观察到的计数的偏度和零膨胀，并同时调整附近单元的现有空间自相关，该方法有助于估计GLCM晶格内部的纹理图案分布。该技术适用于通过CT扫描和不使用造影剂而从CT扫描获得的肾上腺病变的群集图像。我们通过研究其与病理学诊断的对应关系，进一步评估所得亚型是否以临床为导向。另外，我们将性能与模拟研究中的癌症放射学中当前使用的一类机器学习方法进行了比较。我们通过研究其与病理学诊断的对应关系，进一步评估了所产生的亚型是否以临床为导向。另外，我们将性能与模拟研究中的癌症放射学中当前使用的一类机器学习方法进行了比较。我们通过研究其与病理学诊断的对应关系，进一步评估了所产生的亚型是否以临床为导向。此外，我们将性能与目前在癌症放射学中使用的一类机器学习方法进行了模拟研究进行比较。