Robustness of Radiomics in Pre-Surgical Computer Tomography of Non-Small-Cell Lung Cancer
<b>Background:</b> Radiomic features are increasingly used in CT of NSCLC. However, their robustness with respect to segmentation variability has not yet been demonstrated. The aim of this study was to assess radiomic features agreement across three kinds of segmentation. <b>Method...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-12-01
|
| Series: | Journal of Personalized Medicine |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-4426/13/1/83 |
| _version_ | 1797440054515204096 |
|---|---|
| author | Maria Paola Belfiore Mario Sansone Riccardo Monti Stefano Marrone Roberta Fusco Valerio Nardone Roberto Grassi Alfonso Reginelli |
| author_facet | Maria Paola Belfiore Mario Sansone Riccardo Monti Stefano Marrone Roberta Fusco Valerio Nardone Roberto Grassi Alfonso Reginelli |
| author_sort | Maria Paola Belfiore |
| collection | DOAJ |
| description | <b>Background:</b> Radiomic features are increasingly used in CT of NSCLC. However, their robustness with respect to segmentation variability has not yet been demonstrated. The aim of this study was to assess radiomic features agreement across three kinds of segmentation. <b>Methods</b>: We retrospectively included 48 patients suffering from NSCLC who underwent pre-surgery CT. Two expert radiologists in consensus manually delineated three 3D-ROIs on each patient. To assess robustness for each feature, the intra-class correlation coefficient (ICC) across segmentations was evaluated. The ‘sensitivity’ of ICC upon some parameters affecting features computation (such as bin-width for first-order features and pixel-distances for second-order features) was also evaluated. Moreover, an assessment with respect to interpolator and isotropic resolution was also performed. <b>Results</b>: Our results indicate that ‘shape’ features tend to have excellent agreement (ICC > 0.9) across segmentations; moreover, they have approximately zero sensitivity to other parameters. ‘First-order’ features are in general sensitive to parameters variation; however, a few of them showed excellent agreement and low sensitivity (below 0.1) with respect to bin-width and pixel-distance. Similarly, a few second-order features showed excellent agreement and low sensitivity. <b>Conclusions</b>: Our results suggest that a limited number of radiomic features can achieve a high level of reproducibility in CT of NSCLC. |
| first_indexed | 2024-03-09T12:02:37Z |
| format | Article |
| id | doaj.art-4b3132d3e04945e39bad17ac670f2716 |
| institution | Directory Open Access Journal |
| issn | 2075-4426 |
| language | English |
| last_indexed | 2024-03-09T12:02:37Z |
| publishDate | 2022-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Journal of Personalized Medicine |
| spelling | doaj.art-4b3132d3e04945e39bad17ac670f27162023-11-30T23:01:54ZengMDPI AGJournal of Personalized Medicine2075-44262022-12-011318310.3390/jpm13010083Robustness of Radiomics in Pre-Surgical Computer Tomography of Non-Small-Cell Lung CancerMaria Paola Belfiore0Mario Sansone1Riccardo Monti2Stefano Marrone3Roberta Fusco4Valerio Nardone5Roberto Grassi6Alfonso Reginelli7Departement of Precision Medicine, Campania University “Luigi Vanvitelli”, 80138 Naples, ItalyDepartment of Electrical Engineering and Information Technology, University ‘Federico II’, 80131 Naples, ItalyDepartement of Precision Medicine, Campania University “Luigi Vanvitelli”, 80138 Naples, ItalyDepartment of Electrical Engineering and Information Technology, University ‘Federico II’, 80131 Naples, ItalyDepartment of Research & Development IGEA Span, 80013 Carpi, ItalyDepartement of Precision Medicine, Campania University “Luigi Vanvitelli”, 80138 Naples, ItalyDepartement of Precision Medicine, Campania University “Luigi Vanvitelli”, 80138 Naples, ItalyDepartement of Precision Medicine, Campania University “Luigi Vanvitelli”, 80138 Naples, Italy<b>Background:</b> Radiomic features are increasingly used in CT of NSCLC. However, their robustness with respect to segmentation variability has not yet been demonstrated. The aim of this study was to assess radiomic features agreement across three kinds of segmentation. <b>Methods</b>: We retrospectively included 48 patients suffering from NSCLC who underwent pre-surgery CT. Two expert radiologists in consensus manually delineated three 3D-ROIs on each patient. To assess robustness for each feature, the intra-class correlation coefficient (ICC) across segmentations was evaluated. The ‘sensitivity’ of ICC upon some parameters affecting features computation (such as bin-width for first-order features and pixel-distances for second-order features) was also evaluated. Moreover, an assessment with respect to interpolator and isotropic resolution was also performed. <b>Results</b>: Our results indicate that ‘shape’ features tend to have excellent agreement (ICC > 0.9) across segmentations; moreover, they have approximately zero sensitivity to other parameters. ‘First-order’ features are in general sensitive to parameters variation; however, a few of them showed excellent agreement and low sensitivity (below 0.1) with respect to bin-width and pixel-distance. Similarly, a few second-order features showed excellent agreement and low sensitivity. <b>Conclusions</b>: Our results suggest that a limited number of radiomic features can achieve a high level of reproducibility in CT of NSCLC.https://www.mdpi.com/2075-4426/13/1/83radiomiclung cancerradiogenomicCTtexture analysis |
| spellingShingle | Maria Paola Belfiore Mario Sansone Riccardo Monti Stefano Marrone Roberta Fusco Valerio Nardone Roberto Grassi Alfonso Reginelli Robustness of Radiomics in Pre-Surgical Computer Tomography of Non-Small-Cell Lung Cancer Journal of Personalized Medicine radiomic lung cancer radiogenomic CT texture analysis |
| title | Robustness of Radiomics in Pre-Surgical Computer Tomography of Non-Small-Cell Lung Cancer |
| title_full | Robustness of Radiomics in Pre-Surgical Computer Tomography of Non-Small-Cell Lung Cancer |
| title_fullStr | Robustness of Radiomics in Pre-Surgical Computer Tomography of Non-Small-Cell Lung Cancer |
| title_full_unstemmed | Robustness of Radiomics in Pre-Surgical Computer Tomography of Non-Small-Cell Lung Cancer |
| title_short | Robustness of Radiomics in Pre-Surgical Computer Tomography of Non-Small-Cell Lung Cancer |
| title_sort | robustness of radiomics in pre surgical computer tomography of non small cell lung cancer |
| topic | radiomic lung cancer radiogenomic CT texture analysis |
| url | https://www.mdpi.com/2075-4426/13/1/83 |
| work_keys_str_mv | AT mariapaolabelfiore robustnessofradiomicsinpresurgicalcomputertomographyofnonsmallcelllungcancer AT mariosansone robustnessofradiomicsinpresurgicalcomputertomographyofnonsmallcelllungcancer AT riccardomonti robustnessofradiomicsinpresurgicalcomputertomographyofnonsmallcelllungcancer AT stefanomarrone robustnessofradiomicsinpresurgicalcomputertomographyofnonsmallcelllungcancer AT robertafusco robustnessofradiomicsinpresurgicalcomputertomographyofnonsmallcelllungcancer AT valerionardone robustnessofradiomicsinpresurgicalcomputertomographyofnonsmallcelllungcancer AT robertograssi robustnessofradiomicsinpresurgicalcomputertomographyofnonsmallcelllungcancer AT alfonsoreginelli robustnessofradiomicsinpresurgicalcomputertomographyofnonsmallcelllungcancer |