PET textural features stability and pattern discrimination power for radiomics analysis: An “ad-hoc” phantoms study

Presotto, Luca; Bettinardi, Valentino; Bernardi, Elisabetta De; Belli, Maria Luisa; Cattaneo, Giovanni Mauro; Broggi, S.; Fiorino, Claudio

doi:10.1016/j.ejmp.2018.05.024

Cited by 36 publications

(26 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In these calculations, we used the full range of activity concentrations delivered by the PET/CT system for quantizing In cases where the greyscale mapping was restricted to a fixed bin size scheme with a lower bound on zero (c.f. S1 Text) as suggested by Presotto et al [26], variability inflated to four…”

Section: Plos Onementioning

confidence: 66%

“…Furthermore, stability occurs at lower exposures with greyscale mapping (Figs 7 and S5). However, this probably only holds true in a perfectly homogenous VOI, such as used in this work: Figs 4 and S5 and S7, and the work of Presotto et al [26] hint that the quantization distance between grey level groups can introduce new variability. Such a situation might occur were lesions of similar texture are found within differing background ACs.…”

Section: Plos Onementioning

confidence: 82%

“…It is therefore unsurprising that studies without greyscale mapped GLCMs report high feature variability [21], whereas studies using greyscale mapping yielded more reproducible results [19,24]. Furthermore, when using a fixed bin size scheme with the lower bound on zero [26], we found increased variability in the absence of exposures with stable plateaus (S7 Fig). However, neither feature variability (Fig 6) nor the occurrence of stable plateaus (Fig 7), were affected much by GLCM bit size.…”

Section: Plos Onementioning

confidence: 85%

“…However, the contribution of image count statistics, a known driver of variability in imaging, remains largely overlooked. Although, some studies show effects of noise on image texture [25][26][27], others have reported insignificant effects of image statistics on textural features [28,29]. In fact, there has hitherto be no systematic study of the progression of textural feature values with changing photon count.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Towards guidelines to harmonize textural features in PET: Haralick textural features vary with image noise, but exposure-invariant domains enable comparable PET radiomics

et al. 2020

View full text Add to dashboard Cite

PurposeImage texture is increasingly used to discriminate tissues and lesions in PET/CT. For quantification or in computer-aided diagnosis, textural feature analysis must produce robust and comparable values. Because statistical feature values depend on image count statistics, we investigated in depth the stability of Haralick features values as functions of acquisition duration, and for common image resolutions and reconstructions. MethodsA homogeneous cylindrical phantom containing 9.6 kBq/ml Ge-68 was repeatedly imaged on a Siemens Biograph mCT, with acquisition durations ranging from three seconds to three hours. Images with 1.5, 2, and 4 mm isometrically spaced voxels were reconstructed with filtered back-projection (FBP), ordered subset expectation maximization (OSEM), and the Siemens TrueX algorithm. We analysed Haralick features derived from differently quantized (3 to 8-bit) grey level co-occurrence matrices (GLCMs) as functions of exposure E, which we defined as the product of activity concentration in a volume of interest (VOI) and acquisition duration. The VOI was a 50 mm wide cube at the centre of the phantom. Feature stability was defined for df/dE ! 0. ResultsThe most stable feature values occurred in low resolution FBPs, whereas some feature values from 1.5 mm TrueX reconstructions ranged over two orders of magnitude. Within the same reconstructions, most feature value-exposure curves reached stable plateaus at similar exposures, regardless of GLCM quantization. With 8-bit GLCM, median time to stability was 16 s and 22 s for FBPs, 18 s and 125 s for OSEM, and 23 s, 45 s, and 76 s for PSF

show abstract

Section: Plos Onementioning

confidence: 66%

Section: Plos Onementioning

confidence: 82%

Section: Plos Onementioning

confidence: 85%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Towards guidelines to harmonize textural features in PET: Haralick textural features vary with image noise, but exposure-invariant domains enable comparable PET radiomics

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Although many factors influence the SUV, a study investigating the stability of FDG-PET radiomics features showed that 71 % of the features were test-retest stable and 91 % were inter-observer stable [74]. Another study showed that for PET radiomics the fixedbin-number discretization method resulted in more stable SUV values than the fixed-binsize discretization method [75]. For radiomics the segmentation process can have a large influence.…”

Section: Influence Of Pet Reconstruction Algorithmsmentioning

confidence: 99%

Image features for the future in stage IV non-small cell lung cancer

Jong¹,

Evelyn²

View full text Add to dashboard Cite

show abstract

Physical imaging phantoms for simulation of tumor heterogeneity in PET, CT, and MRI: An overview of existing designs

2020

View full text Add to dashboard Cite

Background: In oncology, lesion characterization is essential for tumor grading, treatment planning, and follow-up of cancer patients. Hybrid imaging systems, such as Single Photon Emission Computed Tomography (SPECT)/CT, Positron Emission Tomography (PET)/CT, or PET/magnetic resonance imaging (MRI), play an essential role for the noninvasive quantification of tumor characteristics. However, most of the existing approaches are challenged by intra-and intertumor heterogeneity. Novel quantitative imaging parameters that can be derived from textural feature analysis (as part of radiomics) are promising complements for improved characterization of tumor heterogeneity, thus, supporting clinically relevant implementations of personalized medicine concepts. Nevertheless, establishing new quantitative parameters for tumor characterization requires the use of standardized imaging objects to test the reliability of results prior to their implementation in patient studies. Methods: In this review, we summarize existing reports on heterogeneous phantoms with a focus on simulating tumor heterogeneity. We discuss the techniques, materials, advantages, and limitations of the existing phantoms for PET, CT, and MR imaging modalities. Conclusions: Finally, we outline the future directions and requirements for the design of cross modality imaging phantoms.

show abstract

PET textural features stability and pattern discrimination power for radiomics analysis: An “ad-hoc” phantoms study

Cited by 36 publications

References 42 publications

Towards guidelines to harmonize textural features in PET: Haralick textural features vary with image noise, but exposure-invariant domains enable comparable PET radiomics

Towards guidelines to harmonize textural features in PET: Haralick textural features vary with image noise, but exposure-invariant domains enable comparable PET radiomics

Image features for the future in stage IV non-small cell lung cancer

Physical imaging phantoms for simulation of tumor heterogeneity in PET, CT, and MRI: An overview of existing designs

Contact Info

Product

Resources

About