Evaluation of radiomics feature stability in abdominal monoenergetic photon counting CT reconstructions

Abstract

Feature stability and standardization remain challenges that impede the clinical implementation of radiomics. This study investigates the potential of spectral reconstructions from photon-counting computed tomography (PCCT) regarding organ-specific radiomics feature stability. Abdominal portal-venous phase PCCT scans of 10 patients in virtual monoenergetic (VM) (keV 40–120 in steps of 10), polyenergetic, virtual non-contrast (VNC), and iodine maps were acquired. Two 2D and 3D segmentations measuring 1 and 2 cm in diameter of the liver, lung, spleen, psoas muscle, subcutaneous fat, and air were obtained for spectral reconstructions. Radiomics features were extracted with pyradiomics. The calculation of feature-specific intraclass correlation coefficients (ICC) was performed by comparing all segmentation approaches and organs. Feature-wise and organ-wise correlations were evaluated. Segmentation-resegmentation stability was evaluated by concordance correlation coefficient (CCC). Compared to non-VM, VM-reconstruction features tended to be more stable. For VM reconstructions, 3D 2 cm segmentation showed the highest average ICC with 0.63. Based on a criterion of ≥ 3 stable organs and an ICC of ≥ 0.75, 12—mainly non-first-order features—are shown to be stable between the VM reconstructions. In a segmentation-resegmentation analysis in 3D 2 cm, three features were identified as stable based on a CCC of > 0.6 in ≥ 3 organs in ≥ 6 VM reconstructions. Certain radiomics features vary between monoenergetic reconstructions and depend on the ROI size. Feature stability was also shown to differ between different organs. Yet, glcm_JointEntropy, gldm_GrayLevelNonUniformity, and firstorder_Entropy could be identified as features that could be interpreted as energy-independent and segmentation-resegmentation stable in this PCCT collective. PCCT may support radiomics feature standardization and comparability between sites.

Photon-counting detector CT-based virtual monoenergetic reconstructions: repeatability and reproducibility of radiomics features of an organic phantom and human myocardium

“…Additionally, different tissues had different features with the best reproducibility. Interestingly, the twelve radiomic features with the highest agreement found in the study by Tharmaseelan et al [ 21 ] did not have high reproducibility in our study. Perhaps this difference was due to different acquisition settings or content of the phantom/tissue being imaged.…”

“…Conversely, when image settings such as kernel or spatial resolution change, there is a fundamental decrease in radiomic feature reproducibility. For example, Dunning et al [ 35 ] demonstrated that 13 of 14 relevant radiomic features differed by more than 50% between PCD- and EID-CT. Further, Tharmaseelan et al [ 21 ] demonstrated that VMI (40–120 keV in 10 steps) has an important impact on radiomic feature stability in various abdominal tissue (liver, lung, spleen, psoas muscle and subcutaneous fat). Of the 93 features analyzed in their study, only 12 had ICC values ≥ 0.75 in three or more organs.…”

“…From previous literature, twelve radiomic features have demonstrated stability between VMIs ranging from 40 to 120 keV on PCD-CT scans between different organs (firstorder_Entropy, glcm_Differenceentropy, glcm_Jointentropy, gldm_DependenceEntropy, gldm_DependenceNonUniformity, gldm_DependenceNonUniformityNormalized, gldm_GrayLevelNonUniformity, glrlm_GrayLevelNonUniformity, glrlm_RunLengthNonUniformity, glrlm_RunLengthNonUniformityNormalized, glrlm_ShortRunEmphasis, ngtdm_Coarseness) [ 21 ]. We also specifically examined these preselected features according to our phantom study via intraclass correlation coefficients (ICC) calculations.…”

Technical Basics and Clinical Benefits of Photon-Counting CT

“…For example, it has been investigated whether PCD-CT might not allow the detection of textural changes in periaortic adipose tissue in relation to the presence of local aortic calcification because of its better spatial resolution and signal-to-noise ratio, and thus possibly provide a biomarker for the development of atherosclerosis. 88 The stability of radiomics features in abdominal VMIs 89 and in the myocardium 90 has been investigated, and it could be shown that some texture alterations of the left ventricular myocardium might be associated with the severity of coronary artery calcification estimated by the Agatston score. 91 We are here only at the beginning of a development whose real potential will become apparent in further studies.…”

Pericoronary radiomics texture features associated with hypercholesterolemia on a photon-counting-CT

“…Nevertheless, other studies in our institution have focused on the impact of the detector type (EICT vs. PCCT) on radiomics analysis of the left ventricular myocardium, outlining a possibly deeper insight into the texture by PCCT ( 30 ). The capability of PCCT of constructing monochromatic reconstructions with different keV could further improve feature stability, especially regarding second order texture features and future studies should focus on exploring the clinical applicability of this aspect ( 31 ). The high stability of radiomics features when using PCCT could recently be demonstrated with a phantom model, pointing towards a probable improvement in comparability through the implementation of PCCT ( 8 ).…”