Reproducibility of CT-based radiomic features against image resampling and perturbations for tumour and healthy kidney in renal cancer patients

Abstract Computed Tomography (CT) is widely used in oncology for morphological evaluation and diagnosis, commonly through visual assessments, often exploiting semi-automatic tools as well. Well-established automatic methods for quantitative imaging offer the opportunity to enrich the radiologist int...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Margherita Mottola, Stephan Ursprung, Leonardo Rundo, Lorena Escudero Sanchez, Tobias Klatte, Iosif Mendichovszky, Grant D Stewart, Evis Sala, Alessandro Bevilacqua
Formato: article
Lenguaje:EN
Publicado: Nature Portfolio 2021
Materias:
R
Q
Acceso en línea:https://doaj.org/article/c0efd4e12f234fa7abb5a8e2518d129a
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:c0efd4e12f234fa7abb5a8e2518d129a
record_format dspace
spelling oai:doaj.org-article:c0efd4e12f234fa7abb5a8e2518d129a2021-12-02T18:24:53ZReproducibility of CT-based radiomic features against image resampling and perturbations for tumour and healthy kidney in renal cancer patients10.1038/s41598-021-90985-y2045-2322https://doaj.org/article/c0efd4e12f234fa7abb5a8e2518d129a2021-06-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-90985-yhttps://doaj.org/toc/2045-2322Abstract Computed Tomography (CT) is widely used in oncology for morphological evaluation and diagnosis, commonly through visual assessments, often exploiting semi-automatic tools as well. Well-established automatic methods for quantitative imaging offer the opportunity to enrich the radiologist interpretation with a large number of radiomic features, which need to be highly reproducible to be used reliably in clinical practice. This study investigates feature reproducibility against noise, varying resolutions and segmentations (achieved by perturbing the regions of interest), in a CT dataset with heterogeneous voxel size of 98 renal cell carcinomas (RCCs) and 93 contralateral normal kidneys (CK). In particular, first order (FO) and second order texture features based on both 2D and 3D grey level co-occurrence matrices (GLCMs) were considered. Moreover, this study carries out a comparative analysis of three of the most commonly used interpolation methods, which need to be selected before any resampling procedure. Results showed that the Lanczos interpolation is the most effective at preserving original information in resampling, where the median slice resolution coupled with the native slice spacing allows the best reproducibility, with 94.6% and 87.7% of features, in RCC and CK, respectively. GLCMs show their maximum reproducibility when used at short distances.Margherita MottolaStephan UrsprungLeonardo RundoLorena Escudero SanchezTobias KlatteIosif MendichovszkyGrant D StewartEvis SalaAlessandro BevilacquaNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-11 (2021)
institution DOAJ
collection DOAJ
language EN
topic Medicine
R
Science
Q
spellingShingle Medicine
R
Science
Q
Margherita Mottola
Stephan Ursprung
Leonardo Rundo
Lorena Escudero Sanchez
Tobias Klatte
Iosif Mendichovszky
Grant D Stewart
Evis Sala
Alessandro Bevilacqua
Reproducibility of CT-based radiomic features against image resampling and perturbations for tumour and healthy kidney in renal cancer patients
description Abstract Computed Tomography (CT) is widely used in oncology for morphological evaluation and diagnosis, commonly through visual assessments, often exploiting semi-automatic tools as well. Well-established automatic methods for quantitative imaging offer the opportunity to enrich the radiologist interpretation with a large number of radiomic features, which need to be highly reproducible to be used reliably in clinical practice. This study investigates feature reproducibility against noise, varying resolutions and segmentations (achieved by perturbing the regions of interest), in a CT dataset with heterogeneous voxel size of 98 renal cell carcinomas (RCCs) and 93 contralateral normal kidneys (CK). In particular, first order (FO) and second order texture features based on both 2D and 3D grey level co-occurrence matrices (GLCMs) were considered. Moreover, this study carries out a comparative analysis of three of the most commonly used interpolation methods, which need to be selected before any resampling procedure. Results showed that the Lanczos interpolation is the most effective at preserving original information in resampling, where the median slice resolution coupled with the native slice spacing allows the best reproducibility, with 94.6% and 87.7% of features, in RCC and CK, respectively. GLCMs show their maximum reproducibility when used at short distances.
format article
author Margherita Mottola
Stephan Ursprung
Leonardo Rundo
Lorena Escudero Sanchez
Tobias Klatte
Iosif Mendichovszky
Grant D Stewart
Evis Sala
Alessandro Bevilacqua
author_facet Margherita Mottola
Stephan Ursprung
Leonardo Rundo
Lorena Escudero Sanchez
Tobias Klatte
Iosif Mendichovszky
Grant D Stewart
Evis Sala
Alessandro Bevilacqua
author_sort Margherita Mottola
title Reproducibility of CT-based radiomic features against image resampling and perturbations for tumour and healthy kidney in renal cancer patients
title_short Reproducibility of CT-based radiomic features against image resampling and perturbations for tumour and healthy kidney in renal cancer patients
title_full Reproducibility of CT-based radiomic features against image resampling and perturbations for tumour and healthy kidney in renal cancer patients
title_fullStr Reproducibility of CT-based radiomic features against image resampling and perturbations for tumour and healthy kidney in renal cancer patients
title_full_unstemmed Reproducibility of CT-based radiomic features against image resampling and perturbations for tumour and healthy kidney in renal cancer patients
title_sort reproducibility of ct-based radiomic features against image resampling and perturbations for tumour and healthy kidney in renal cancer patients
publisher Nature Portfolio
publishDate 2021
url https://doaj.org/article/c0efd4e12f234fa7abb5a8e2518d129a
work_keys_str_mv AT margheritamottola reproducibilityofctbasedradiomicfeaturesagainstimageresamplingandperturbationsfortumourandhealthykidneyinrenalcancerpatients
AT stephanursprung reproducibilityofctbasedradiomicfeaturesagainstimageresamplingandperturbationsfortumourandhealthykidneyinrenalcancerpatients
AT leonardorundo reproducibilityofctbasedradiomicfeaturesagainstimageresamplingandperturbationsfortumourandhealthykidneyinrenalcancerpatients
AT lorenaescuderosanchez reproducibilityofctbasedradiomicfeaturesagainstimageresamplingandperturbationsfortumourandhealthykidneyinrenalcancerpatients
AT tobiasklatte reproducibilityofctbasedradiomicfeaturesagainstimageresamplingandperturbationsfortumourandhealthykidneyinrenalcancerpatients
AT iosifmendichovszky reproducibilityofctbasedradiomicfeaturesagainstimageresamplingandperturbationsfortumourandhealthykidneyinrenalcancerpatients
AT grantdstewart reproducibilityofctbasedradiomicfeaturesagainstimageresamplingandperturbationsfortumourandhealthykidneyinrenalcancerpatients
AT evissala reproducibilityofctbasedradiomicfeaturesagainstimageresamplingandperturbationsfortumourandhealthykidneyinrenalcancerpatients
AT alessandrobevilacqua reproducibilityofctbasedradiomicfeaturesagainstimageresamplingandperturbationsfortumourandhealthykidneyinrenalcancerpatients
_version_ 1718378120899198976