Impact of contouring variability on oncological PET radiomics features in the lung

F Yang,G Simpson,N Dogan,J Ford,L Young,L Wang

doi:10.1038/s41598-019-57171-7

F Yang, G Simpson + Show 4 more

Open Access

https://doi.org/10.1038/s41598-019-57171-7

Copy DOI

Abstract

Radiomics features extracted from oncological PET images are currently under intense scrutiny within the context of risk stratification for a variety of cancers. However, the lack of robustness assessment poses problems for their application across institutions and for broader patient populations. The objective of the current study was to examine the extent to which radiomics parameters from oncological PET vary in response to manual contouring variability in lung cancer. Imaging data employed in the study consisted of 26 PET scans with lesions in the lung being created through the use of an anthropomorphic phantom in conjunction with Monte Carlo simulations. From each of the simulated lesions, 25 radiomics features related to the gray-level co-occurrence matrices (GLCOM), gray-level size zone matrices (GLSZM), and gray-level neighborhood difference matrices (GLNDM) were extracted from ground truth contour and from manual contours provided by 10 raters in regard to four intensity discretization schemes with number of gray levels of 32, 64, 128, and 256, respectively. The impact of interrater variability in tumor delineation upon the agreement between raters on radiomics features was examined via interclass correlation and leave-p-out assessment. Only weak and moderate correlations were found between segmentation accuracy as measured by the Dice coefficient and percent feature error from ground truth for the vast majority of the features being examined. GLNDM-based texture parameters emerged as the top performing category of radiomcs features in terms of robustness against contouring variability for discretization schemes engaging number of gray levels of 32, 64, and 128 while GLCOM-based parameters stood out for discretization scheme engaging 256 gray levels. How and to what extent interrater reliability of radiomics features vary in response to the number of raters were largely feature-dependent. It was concluded that impact of contouring variability on PET-based radiomics features is present to varying degrees and could be experienced as a barrier to convey PET-based radiomics research to clinical relevance.

Highlights

Feature Energy Contrast Entropy Homogeneity Correlation Variance Dissimilarity Short Zones Emphasis (SZE) Large Zones Emphasis (LZE) Gray-level Non-uniformity (GLN) Zone Size Non-Uniformity (ZSNU) Zone Percentage (ZP) Low Gray-level Zones Emphasis (LGZE) High Gray-level Zones Emphasis (HGZE) Short Zones Low Gray-level Emphasis (SZLGE) Short Zones High Gray-level Emphasis (SZHGE) Large Zones Low Gray-level Emphasis (LZLGE) Large Zones High Gray-level Emphasis (LZHGE) Gray-level Variance Emphasis (GLV) Zone Size Variance Emphasis (ZSV) Coarseness Contrast Busyness Complexity Strength against an array of diverse uncertainties intrinsic to radiomics analysis
As for the variability of contouring volume of interest (VOI) for inclusion to radiomics analysis, only a few existing studies investigated its impact on PET radiomics parameters[15,16], even though contouring variability is commonly recognized as the largest source of error in the RT planning process[17]
By aid of a realistic 3D digital phantom of the thorax in conjunction with a Monte Carlo (MC) based PET imaging simulation package, the purpose of this work was to examine the impact of contouring variability on PET radiomics parameters for lung cancer in a setting with complete knowledge of tumor location, morphology, and intensity distribution

Summary

Introduction

Feature Energy Contrast Entropy Homogeneity Correlation Variance Dissimilarity Short Zones Emphasis (SZE) Large Zones Emphasis (LZE) Gray-level Non-uniformity (GLN) Zone Size Non-Uniformity (ZSNU) Zone Percentage (ZP) Low Gray-level Zones Emphasis (LGZE) High Gray-level Zones Emphasis (HGZE) Short Zones Low Gray-level Emphasis (SZLGE) Short Zones High Gray-level Emphasis (SZHGE) Large Zones Low Gray-level Emphasis (LZLGE) Large Zones High Gray-level Emphasis (LZHGE) Gray-level Variance Emphasis (GLV) Zone Size Variance Emphasis (ZSV) Coarseness Contrast Busyness Complexity Strength against an array of diverse uncertainties intrinsic to radiomics analysis. As for the variability of contouring volume of interest (VOI) for inclusion to radiomics analysis, only a few existing studies investigated its impact on PET radiomics parameters[15,16], even though contouring variability is commonly recognized as the largest source of error in the RT planning process[17]. These studies involving contouring were carried out using clinical PET imaging data with an inherent lack of knowledge about actual tumor volumes. The present work furnished a set of methodological guides exploitable for design and implementation of future investigations endeavoring to translate PET radiomics into clinical relevance

Objectives

Methods

Results

Conclusion