The role of parametric feature maps to correct different volume of interest sizes: an in vivo liver MRI study.
Liver
Magnetic resonance imaging
Radiomics
Reproducibility of results
Software
Journal
European radiology experimental
ISSN: 2509-9280
Titre abrégé: Eur Radiol Exp
Pays: England
ID NLM: 101721752
Informations de publication
Date de publication:
06 09 2023
06 09 2023
Historique:
received:
23
04
2023
accepted:
13
06
2023
medline:
7
9
2023
pubmed:
6
9
2023
entrez:
5
9
2023
Statut:
epublish
Résumé
Different volume of interest (VOI) sizes influence radiomic features. This study examined if translating images into feature maps before feature sampling could compensate for these effects in liver magnetic resonance imaging (MRI). T1- and T2-weighted sequences from three different scanners (two 3-T scanners, one 1.5-T scanner) of 66 patients with normal abdominal MRI were included retrospectively. Three differently sized VOIs (10, 20, and 30 mm in diameter) were drawn in the liver parenchyma (right lobe), excluding adjacent structures. Ninety-three features were extracted conventionally using PyRadiomics. All images were also converted to 93 parametric feature maps using a pretested software. Agreement between the three VOI sizes was assessed with overall concordance correlation coefficients (OCCCs), while OCCCs > 0.85 were rated reproducible. OCCCs were calculated twice: for the VOI sizes of 10, 20, and 30 mm and for those of 20 and 30 mm. When extracted from original images, only 4 out of the 93 features were reproducible across all VOI sizes in T1- and T2-weighted images. When the smallest VOI was excluded, 5 features (T1-weighted) and 7 features (T2-weighted) were reproducible. Extraction from parametric maps increased the number of reproducible features to 9 (T1- and T2-weighted) across all VOIs. Excluding the 10-mm VOI, reproducibility improved to 16 (T1-weighted) and 55 features (T2-weighted). The stability of all other features also increased in feature maps. Translating images into parametric maps before feature extraction improves reproducibility across different VOI sizes in normal liver MRI. The size of the segmented VOI influences the feature quantity of radiomics, while software-based conversion of images into parametric feature maps before feature sampling improves reproducibility across different VOI sizes in MRI of normal liver tissue. • Parametric feature maps can compensate for different VOI sizes. • The effect seems dependent on the VOI sizes and the MRI sequence. • Feature maps can visualize features throughout the entire image stack.
Sections du résumé
BACKGROUND
Different volume of interest (VOI) sizes influence radiomic features. This study examined if translating images into feature maps before feature sampling could compensate for these effects in liver magnetic resonance imaging (MRI).
METHODS
T1- and T2-weighted sequences from three different scanners (two 3-T scanners, one 1.5-T scanner) of 66 patients with normal abdominal MRI were included retrospectively. Three differently sized VOIs (10, 20, and 30 mm in diameter) were drawn in the liver parenchyma (right lobe), excluding adjacent structures. Ninety-three features were extracted conventionally using PyRadiomics. All images were also converted to 93 parametric feature maps using a pretested software. Agreement between the three VOI sizes was assessed with overall concordance correlation coefficients (OCCCs), while OCCCs > 0.85 were rated reproducible. OCCCs were calculated twice: for the VOI sizes of 10, 20, and 30 mm and for those of 20 and 30 mm.
RESULTS
When extracted from original images, only 4 out of the 93 features were reproducible across all VOI sizes in T1- and T2-weighted images. When the smallest VOI was excluded, 5 features (T1-weighted) and 7 features (T2-weighted) were reproducible. Extraction from parametric maps increased the number of reproducible features to 9 (T1- and T2-weighted) across all VOIs. Excluding the 10-mm VOI, reproducibility improved to 16 (T1-weighted) and 55 features (T2-weighted). The stability of all other features also increased in feature maps.
CONCLUSIONS
Translating images into parametric maps before feature extraction improves reproducibility across different VOI sizes in normal liver MRI.
RELEVANCE STATEMENT
The size of the segmented VOI influences the feature quantity of radiomics, while software-based conversion of images into parametric feature maps before feature sampling improves reproducibility across different VOI sizes in MRI of normal liver tissue.
KEY POINTS
• Parametric feature maps can compensate for different VOI sizes. • The effect seems dependent on the VOI sizes and the MRI sequence. • Feature maps can visualize features throughout the entire image stack.
Identifiants
pubmed: 37670193
doi: 10.1186/s41747-023-00362-9
pii: 10.1186/s41747-023-00362-9
pmc: PMC10480134
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
48Informations de copyright
© 2023. European Society of Radiology (ESR).
Références
Ng F, Ganeshan B, Kozarski R, Miles KA, Goh V (2013) Assessment of primary colorectal cancer heterogeneity by using whole-tumor texture analysis: contrast-enhanced CT texture as a biomarker of 5-year survival. Radiology 266:177–184. https://doi.org/10.1148/radiol.12120254
doi: 10.1148/radiol.12120254
pubmed: 23151829
Gillies RJ, Kinahan PE, Hricak H (2016) Radiomics: images are more than pictures, they are data. Radiology 278:563–577. https://doi.org/10.1148/radiol.2015151169
doi: 10.1148/radiol.2015151169
pubmed: 26579733
Becker AS, Schneider MA, Wurnig MC, Wagner M, Clavien PA, Boss A (2018) Radiomics of liver MRI predict metastases in mice. Eur Radiol Exp 2:11. https://doi.org/10.1186/s41747-018-0044-7
doi: 10.1186/s41747-018-0044-7
pubmed: 29882527
pmcid: 5971192
Fradet G, Ayde R, Bottois H et al (2022) Prediction of lipomatous soft tissue malignancy on MRI: comparison between machine learning applied to radiomics and deep learning. Eur Radiol Exp 6:41. https://doi.org/10.1186/s41747-022-00295-9
doi: 10.1186/s41747-022-00295-9
pubmed: 36071368
pmcid: 9452614
Jensen LJ, Kim D, Elgeti T, Steffen IG, Hamm B, Nagel SN (2022) Differentiation of pulmonary lymphoma manifestations and nonlymphoma infiltrates in possible invasive fungal disease using fast T1-weighted magnetic resonance imaging at 3 T comparison of texture analysis, mapping, and signal intensity quotients. J Thorac Imaging 37:80–89. https://doi.org/10.1097/RTI.0000000000000606
doi: 10.1097/RTI.0000000000000606
pubmed: 34269753
Jensen LJ, Rogasch JMM, Kim D et al (2022) CT radiomics to predict deauville score 4 positive and negative Hodgkin lymphoma manifestations. Sci Rep 12:20008. https://doi.org/10.1038/s41598-022-24227-0
doi: 10.1038/s41598-022-24227-0
pubmed: 36411307
pmcid: 9678888
Aerts HJ, Velazquez ER, Leijenaar RT et al (2014) Decoding tumour phenotype by noninvasive imaging using a quantitative radiomics approach. Nat Commun 5:4006. https://doi.org/10.1038/ncomms5006
doi: 10.1038/ncomms5006
pubmed: 24892406
Pinto Dos Santos D, Dietzel M, Baessler B (2021) A decade of radiomics research: are images really data or just patterns in the noise? Eur Radiol 31:1–4. https://doi.org/10.1007/s00330-020-07108-w
doi: 10.1007/s00330-020-07108-w
pubmed: 32767103
Wood ML (2020) Variability and standardization of quantitative imaging. Invest Radiol 55:617–618. https://doi.org/10.1097/RLI.0000000000000667
doi: 10.1097/RLI.0000000000000667
pubmed: 32776768
Hagiwara A, Fujita S, Ohno Y, Aoki S (2020) Variability and standardization of quantitative imaging: monoparametric to multiparametric quantification, radiomics, and artificial Intelligence. Invest Radiol 55:601–616. https://doi.org/10.1097/RLI.0000000000000666
doi: 10.1097/RLI.0000000000000666
pubmed: 32209816
pmcid: 7413678
Rinaldi L, De Angelis SP, Raimondi S et al (2022) Reproducibility of radiomic features in CT images of NSCLC patients: an integrative analysis on the impact of acquisition and reconstruction parameters. Eur Radiol Exp 6:2. https://doi.org/10.1186/s41747-021-00258-6
doi: 10.1186/s41747-021-00258-6
pubmed: 35075539
pmcid: 8786992
Shur J, Blackledge M, D’Arcy J et al (2021) MRI texture feature repeatability and image acquisition factor robustness, a phantom study and in silico study. Eur Radiol Exp 5:2. https://doi.org/10.1186/s41747-020-00199-6
Rizzo S, Botta F, Raimondi S et al (2018) Radiomics: the facts and the challenges of image analysis. Eur Radiol Exp 2:36. https://doi.org/10.1186/s41747-018-0068-z
doi: 10.1186/s41747-018-0068-z
pubmed: 30426318
pmcid: 6234198
Berenguer R, Pastor-Juan MDR, Canales-Vazquez J et al (2018) Radiomics of CT features may be nonreproducible and redundant: influence of CT acquisition parameters. Radiology 288:407–415. https://doi.org/10.1148/radiol.2018172361
doi: 10.1148/radiol.2018172361
pubmed: 29688159
Jensen LJ, Kim D, Elgeti T, Steffen IG, Hamm B, Nagel SN (2021) Stability of radiomic features across different region of interest sizes-a CT and MR phantom study. Tomography 7:238–252. https://doi.org/10.3390/tomography7020022
doi: 10.3390/tomography7020022
pubmed: 34201012
pmcid: 8293351
Jensen LJ, Kim D, Elgeti T, Steffen IG, Hamm B, Nagel SN (2021) Stability of liver radiomics across different 3D ROI sizes-an MRI in vivo study. Tomography 7:866–876. https://doi.org/10.3390/tomography7040073
doi: 10.3390/tomography7040073
pubmed: 34941645
pmcid: 8706942
Roy S, Whitehead TD, Quirk JD et al (2020) Optimal co-clinical radiomics: sensitivity of radiomic features to tumour volume, image noise and resolution in co-clinical T1-weighted and T2-weighted magnetic resonance imaging. EBioMedicine 59:102963. https://doi.org/10.1016/j.ebiom.2020.102963
doi: 10.1016/j.ebiom.2020.102963
pubmed: 32891051
pmcid: 7479492
Traverso A, Kazmierski M, Zhovannik I et al (2020) Machine learning helps identifying volume-confounding effects in radiomics. Phys Med 71:24–30. https://doi.org/10.1016/j.ejmp.2020.02.010
doi: 10.1016/j.ejmp.2020.02.010
pubmed: 32088562
Choi W, Riyahi S, Kligerman SJ, Liu CJ, Mechalakos JG, Lu W (2018) Technical note: identification of CT texture features robust to tumor size variations for normal lung texture analysis. Int J Med Phys Clin Eng Radiat Oncol 7:330–338. https://doi.org/10.4236/ijmpcero.2018.73027
doi: 10.4236/ijmpcero.2018.73027
pubmed: 31131158
pmcid: 6530926
Jensen LJ, Kim D, Elgeti T et al (2022) Enhancing the stability of CT radiomics across different volume of interest sizes using parametric feature maps: a phantom study. Eur Radiol Exp 6:43. https://doi.org/10.1186/s41747-022-00297-7
doi: 10.1186/s41747-022-00297-7
pubmed: 36104519
pmcid: 9474978
Kim D, Jensen LJ, Elgeti T, Steffen IG, Hamm B, Nagel SN (2021) Radiomics for everyone: a new tool simplifies creating parametric maps for the visualization and quantification of radiomics features. Tomography 7:477–487. https://doi.org/10.3390/tomography7030041
doi: 10.3390/tomography7030041
pubmed: 34564303
pmcid: 8482265
Fedorov A, Beichel R, Kalpathy-Cramer J et al (2012) 3D Slicer as an image computing platform for the quantitative imaging network. Magn Reson Imaging 30:1323–1341. https://doi.org/10.1016/j.mri.2012.05.001
doi: 10.1016/j.mri.2012.05.001
pubmed: 22770690
pmcid: 3466397
Serai SD, Obuchowski NA, Venkatesh SK et al (2017) Repeatability of MR elastography of liver: a meta-analysis. Radiology 285:92–100. https://doi.org/10.1148/radiol.2017161398
doi: 10.1148/radiol.2017161398
pubmed: 28530847
PyRadiomics Community (2016) Radiomic features https://pyradiomics.readthedocs.io/en/latest/features.html . Accessed 20 Jan 2023
Zwanenburg A, Vallieres M, Abdalah MA et al (2020) The image biomarker standardization initiative: standardized quantitative radiomics for high-throughput image-based phenotyping. Radiology 295:328–338. https://doi.org/10.1148/radiol.2020191145
doi: 10.1148/radiol.2020191145
pubmed: 32154773
R Development Core Team (2016) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna https://www.R-project.org/ . Accessed 13 Feb 2023
Lin LI (1989) A concordance correlation coefficient to evaluate reproducibility. Biometrics 45:255–268
doi: 10.2307/2532051
pubmed: 2720055
Barnhart HX, Haber M, Song J (2002) Overall concordance correlation coefficient for evaluating agreement among multiple observers. Biometrics 58:1020–1027. https://doi.org/10.1111/j.0006-341x.2002.01020.x
doi: 10.1111/j.0006-341x.2002.01020.x
pubmed: 12495158
Stevenson M, Sergeant E, Heuer C et al (2021) Package “epiR” https://CRAN.R-project.org/package=epiR . Accessed 25 Feb 2023
Santinha J, Bianchini L, Figueiredo M et al (2022) Discrimination of tumor texture based on MRI radiomic features: is there a volume threshold? a phantom study. Appl Sci 12:5465. https://doi.org/10.3390/app12115465
doi: 10.3390/app12115465
Saltybaeva N, Tanadini-Lang S, Vuong D et al (2022) Robustness of radiomic features in magnetic resonance imaging for patients with glioblastoma: multi-center study. Phys Imaging Radiat Oncol 22:131–136. https://doi.org/10.1016/j.phro.2022.05.006
doi: 10.1016/j.phro.2022.05.006
pubmed: 35633866
pmcid: 9130546
Fave X, Zhang L, Yang J et al (2016) Impact of image preprocessing on the volume dependence and prognostic potential of radiomics features in non-small cell lung cancer. Transl Cancer Res 5:349–363
doi: 10.21037/tcr.2016.07.11
Shafiq-Ul-Hassan M, Latifi K, Zhang G, Ullah G, Gillies R, Moros E (2018) Voxel size and gray level normalization of CT radiomic features in lung cancer. Sci Rep 8:10545. https://doi.org/10.1038/s41598-018-28895-9
doi: 10.1038/s41598-018-28895-9
pubmed: 30002441
pmcid: 6043486
Shafiq-ul-Hassan M, Zhang GG, Latifi K et al (2017) Intrinsic dependencies of CT radiomic features on voxel size and number of gray levels. Med phys 44:1050–1062. https://doi.org/10.1002/mp.12123
doi: 10.1002/mp.12123
pubmed: 28112418
pmcid: 5462462
Larue R, van Timmeren JE, de Jong EEC et al (2017) Influence of gray level discretization on radiomic feature stability for different CT scanners, tube currents and slice thicknesses: a comprehensive phantom study. Acta Oncol 56:1544–1553. https://doi.org/10.1080/0284186X.2017.1351624
doi: 10.1080/0284186X.2017.1351624
pubmed: 28885084
Ligero M, Jordi-Ollero O, Bernatowicz K et al (2021) Minimizing acquisition-related radiomics variability by image resampling and batch effect correction to allow for large-scale data analysis. Eur Radiol 31:1460–1470. https://doi.org/10.1007/s00330-020-07174-0
doi: 10.1007/s00330-020-07174-0
pubmed: 32909055
Öllers M, Bosmans G, van Baardwijk A et al (2008) The integration of PET-CT scans from different hospitals into radiotherapy treatment planning. Radiother Oncol 87:142–146. https://doi.org/10.1016/j.radonc.2007.12.025
doi: 10.1016/j.radonc.2007.12.025
pubmed: 18207269
Lu L, Ahmed FS, Akin O et al (2021) Uncontrolled confounders may lead to false or overvalued radiomics signature: a proof of concept using survival analysis in a multicenter cohort of kidney cancer. Front Oncol 11:638185. https://doi.org/10.3389/fonc.2021.638185
doi: 10.3389/fonc.2021.638185
pubmed: 34123789
pmcid: 8191735
Zhang H, Li W, Hu F, Sun Y, Hu T, Tong T (2019) MR texture analysis: potential imaging biomarker for predicting the chemotherapeutic response of patients with colorectal liver metastases. Abdom Radiol (NY) 44:65–71. https://doi.org/10.1007/s00261-018-1682-1
doi: 10.1007/s00261-018-1682-1
pubmed: 29967982
Sim JZT, Hui TCH, Chuah TK, Low HM, Tan CH, Shelat VG (2022) Efficacy of texture analysis of pre-operative magnetic resonance imaging in predicting microvascular invasion in hepatocellular carcinoma. World J Clin Oncol 13:918–928. https://doi.org/10.5306/wjco.v13.i11.918
doi: 10.5306/wjco.v13.i11.918
pubmed: 36483976
pmcid: 9724184
Runge VM, Heverhagen JT (2022) Signal-to-noise ratio in clinical 3 T. the physics of clinical MR taught through images. Springer, Cham; p. 40−41 https://doi.org/10.1007/978-3-030-85413-3
Rai R, Holloway LC, Brink C et al (2020) Multicenter evaluation of MRI-based radiomic features: a phantom study. Med Phys 47:3054–3063. https://doi.org/10.1002/mp.14173
doi: 10.1002/mp.14173
pubmed: 32277703
Ammari S, Pitre-Champagnat S, Dercle L et al (2020) Influence of magnetic field strength on magnetic resonance imaging radiomics features in brain imaging, an in vitro and in vivo Study. Front Oncol 10:541663. https://doi.org/10.3389/fonc.2020.541663
doi: 10.3389/fonc.2020.541663
pubmed: 33552944
Mayerhoefer ME, Szomolanyi P, Jirak D et al (2009) Effects of magnetic resonance image interpolation on the results of texture-based pattern classification: a phantom study. Invest Radiol 44:405–411. https://doi.org/10.1097/RLI.0b013e3181a50a66
doi: 10.1097/RLI.0b013e3181a50a66
pubmed: 19465863