Investigating the Robustness Neighborhood Gray Tone Difference Matrix and Gray Level Co-occurrence Matrix Radiomic Features on Clinical Computed Tomography Systems Using Anthropomorphic Phantoms: Evidence From a Multivendor Study

Abstract

Objective To determine if optimized imaging protocols across multiple computed tomography (CT) vendors could result in reproducible radiomic features calculated from an anthropomorphic phantom. Methods Materials with varying degrees of heterogeneity were placed throughout the lungs of the phantom. Twenty scans of the phantom were acquired on 3 CT manufacturers with chest CT protocols that had optimized protocol parameters. Scans were reconstructed using vendor-specific standards and lung kernels. The Concordance Correlation Coefficient (CCC) was used to calculate reproducibility between features. For features with high CCC values, Bland-Altman analysis was also used to quantify agreement. Results The mean Hounsfield unit (HU) was 32.93 HU (141.7 to −26.5 HU) for the rubber insert and 347.2 HU (−320.9 to −347.7 HU) for the wood insert. Low CCC values < 0.9 were calculated for all features across all scans. Conclusions Radiomic features that are derived from the spatial distribution of voxel intensities should be particularly scrutinized for reproducibility in a multi-vendor environment.

Radiomics as a personalized medicine tool in lung cancer: Separating the hope from the hype

“…Studies assessing the impact of different CT scanners and protocols on radiomic features have shown some features have poor reproducibility [79,80,[84][85][86][87]. Performing phantom studies on different scanners as a quality assurance step may ensure a level of feature consistency [84].…”

“…Image acquisition A Different scanners and acquisition protocols affect feature reproducibility [79][80][81][82][83][84][85][86][87][88][89][90][91] Image phantoms on different scanners to provide baseline [79], establish credibility of scanners and protocols [84], catalogue reproducible features [86,90], model a correction algorithm [89] [98,[110][111][112] Normalization of features to volume [98], bit depth resampling [110], feature redesign [110], more robust statistics to check added value of radiomics signatures [111]. Test re-test H Radiomic features may not be repeatable over multiple measurements [113][114][115], repeatable features are not generalizable to other disease sites [116].…”

CT Texture Analysis Challenges: Influence of Acquisition and Reconstruction Parameters: A Comprehensive Review

“…Mahmood et al [21] used an anthropomorphic lung phantom with shredded rubber and sycamore wood inserts to perform acquisitions on machines from three manufacturers, with a constant voxel size, kVp, pitch and slice thickness. None of the 27 texture features computed with IBEX passed the reproducibility criteria.…”

Effect of tube current on computed tomography radiomic features

“…raised questions about the robustness of image texture features. They imaged an anthropomorphic phantom on scanners from three manufacturers, and no features had CCC values greater than 0.9 24 . A more reliable answer would involve scanning an object using different tube currents and then assessing the impact of this variation on the calculated image features.…”