Impact of segmentation and discretization on radiomic features in 68Ga-DOTA-TOC PET/CT images of neuroendocrine tumor

Virginia Liberini; Bruno De Santi; Osvaldo Rampado; Elena Gallio; Beatrice Dionisi; Francesco Ceci; Giulia Polverari; Philippe Thuillier; Filippo Molinari; Désirée Deandreis

doi:10.1186/s40658-021-00367-6

Impact of segmentation and discretization on radiomic features in ⁶⁸Ga-DOTA-TOC PET/CT images of neuroendocrine tumor

Virginia Liberini^*, Bruno De Santi, Osvaldo Rampado, Elena Gallio, Beatrice Dionisi, Francesco Ceci, Giulia Polverari, Philippe Thuillier, Filippo Molinari, Désirée Deandreis

^*Corresponding author for this work

Research output: Contribution to journal › Article › Academic › peer-review

19 Citations (Scopus)

52 Downloads (Pure)

Abstract

Objective: To identify the impact of segmentation methods and intensity discretization on radiomic features (RFs) extraction from ⁶⁸Ga-DOTA-TOC PET images in patients with neuroendocrine tumors. Methods: Forty-nine patients were retrospectively analyzed. Tumor contouring was performed manually by four different operators and with a semi-automatic edge-based segmentation (SAEB) algorithm. Three SUV_max fixed thresholds (20, 30, 40%) were applied. Fifty-one RFs were extracted applying two different intensity rescale factors for gray-level discretization: one absolute (AR60 = SUV from 0 to 60) and one relative (RR = min-max of the VOI SUV). Dice similarity coefficient (DSC) was calculated to quantify segmentation agreement between different segmentation methods. The impact of segmentation and discretization on RFs was assessed by intra-class correlation coefficients (ICC) and the coefficient of variance (COV^L). The RFs’ correlation with volume and SUV_max was analyzed by calculating Pearson’s correlation coefficients. Results: DSC mean value was 0.75 ± 0.11 (0.45–0.92) between SAEB and operators and 0.78 ± 0.09 (0.36–0.97), among the four manual segmentations. The study showed high robustness (ICC > 0.9): (a) in 64.7% of RFs for segmentation methods using AR60, improved by applying SUV_max threshold of 40% (86.5%); (b) in 50.9% of RFs for different SUV_max thresholds using AR60; and (c) in 37% of RFs for discretization settings using different segmentation methods. Several RFs were not correlated with volume and SUV_max. Conclusions: RFs robustness to manual segmentation resulted higher in NET ⁶⁸Ga-DOTA-TOC images compared to ¹⁸F-FDG PET/CT images. Forty percent SUV_max thresholds yield superior RFs stability among operators, however leading to a possible loss of biological information. SAEB segmentation appears to be an optimal alternative to manual segmentation, but further validations are needed. Finally, discretization settings highly impacted on RFs robustness and should always be stated.

Original language	English
Article number	21
Journal	EJNMMI physics
Volume	8
Issue number	1
Early online date	27 Feb 2021
DOIs	https://doi.org/10.1186/s40658-021-00367-6
Publication status	Published - Dec 2021
Externally published	Yes

Keywords

Ga-DOTATOC PET/CT
Neuroendocrine tumor
Radiomics
Robustness
Semi-automatic segmentation
Texture analysis

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1186/s40658-021-00367-6Licence: CC BY

Liberini-2021-Impact-of-segmentation-and-discretiFinal published version, 1.88 MBLicence: CC BY

Cite this

@article{54a50de3e7bb44e0a9ce2e879381921f,

title = "Impact of segmentation and discretization on radiomic features in 68Ga-DOTA-TOC PET/CT images of neuroendocrine tumor",

abstract = "Objective: To identify the impact of segmentation methods and intensity discretization on radiomic features (RFs) extraction from 68Ga-DOTA-TOC PET images in patients with neuroendocrine tumors. Methods: Forty-nine patients were retrospectively analyzed. Tumor contouring was performed manually by four different operators and with a semi-automatic edge-based segmentation (SAEB) algorithm. Three SUVmax fixed thresholds (20, 30, 40%) were applied. Fifty-one RFs were extracted applying two different intensity rescale factors for gray-level discretization: one absolute (AR60 = SUV from 0 to 60) and one relative (RR = min-max of the VOI SUV). Dice similarity coefficient (DSC) was calculated to quantify segmentation agreement between different segmentation methods. The impact of segmentation and discretization on RFs was assessed by intra-class correlation coefficients (ICC) and the coefficient of variance (COVL). The RFs{\textquoteright} correlation with volume and SUVmax was analyzed by calculating Pearson{\textquoteright}s correlation coefficients. Results: DSC mean value was 0.75 ± 0.11 (0.45–0.92) between SAEB and operators and 0.78 ± 0.09 (0.36–0.97), among the four manual segmentations. The study showed high robustness (ICC > 0.9): (a) in 64.7% of RFs for segmentation methods using AR60, improved by applying SUVmax threshold of 40% (86.5%); (b) in 50.9% of RFs for different SUVmax thresholds using AR60; and (c) in 37% of RFs for discretization settings using different segmentation methods. Several RFs were not correlated with volume and SUVmax. Conclusions: RFs robustness to manual segmentation resulted higher in NET 68Ga-DOTA-TOC images compared to 18F-FDG PET/CT images. Forty percent SUVmax thresholds yield superior RFs stability among operators, however leading to a possible loss of biological information. SAEB segmentation appears to be an optimal alternative to manual segmentation, but further validations are needed. Finally, discretization settings highly impacted on RFs robustness and should always be stated.",

keywords = "Ga-DOTATOC PET/CT, Neuroendocrine tumor, Radiomics, Robustness, Semi-automatic segmentation, Texture analysis",

author = "Virginia Liberini and {De Santi}, Bruno and Osvaldo Rampado and Elena Gallio and Beatrice Dionisi and Francesco Ceci and Giulia Polverari and Philippe Thuillier and Filippo Molinari and D{\'e}sir{\'e}e Deandreis",

note = "Funding Information: Nuclear Medicine of A.O.U. Citt{\`a} della Salute e della Scienza di Torino, University of Turin, is supported by a grant from CRT Foundation (Fondazione Cassa di Risparmio di Torino) for the project “Radiomics in the Study of Neuroendocrine Tumors in Nuclear Medicine: Innovative Method for the Study of Tumor Heterogeneity”. No further specific grants from funding agencies in the public, commercial, or not-for-profit sectors were received for this study. Publisher Copyright: {\textcopyright} 2021, The Author(s).",

year = "2021",

month = dec,

doi = "10.1186/s40658-021-00367-6",

language = "English",

volume = "8",

journal = "EJNMMI physics",

issn = "2197-7364",

publisher = "Springer",

number = "1",

}

TY - JOUR

T1 - Impact of segmentation and discretization on radiomic features in 68Ga-DOTA-TOC PET/CT images of neuroendocrine tumor

AU - Liberini, Virginia

AU - De Santi, Bruno

AU - Rampado, Osvaldo

AU - Gallio, Elena

AU - Dionisi, Beatrice

AU - Ceci, Francesco

AU - Polverari, Giulia

AU - Thuillier, Philippe

AU - Molinari, Filippo

AU - Deandreis, Désirée

N1 - Funding Information: Nuclear Medicine of A.O.U. Città della Salute e della Scienza di Torino, University of Turin, is supported by a grant from CRT Foundation (Fondazione Cassa di Risparmio di Torino) for the project “Radiomics in the Study of Neuroendocrine Tumors in Nuclear Medicine: Innovative Method for the Study of Tumor Heterogeneity”. No further specific grants from funding agencies in the public, commercial, or not-for-profit sectors were received for this study. Publisher Copyright: © 2021, The Author(s).

PY - 2021/12

Y1 - 2021/12

N2 - Objective: To identify the impact of segmentation methods and intensity discretization on radiomic features (RFs) extraction from 68Ga-DOTA-TOC PET images in patients with neuroendocrine tumors. Methods: Forty-nine patients were retrospectively analyzed. Tumor contouring was performed manually by four different operators and with a semi-automatic edge-based segmentation (SAEB) algorithm. Three SUVmax fixed thresholds (20, 30, 40%) were applied. Fifty-one RFs were extracted applying two different intensity rescale factors for gray-level discretization: one absolute (AR60 = SUV from 0 to 60) and one relative (RR = min-max of the VOI SUV). Dice similarity coefficient (DSC) was calculated to quantify segmentation agreement between different segmentation methods. The impact of segmentation and discretization on RFs was assessed by intra-class correlation coefficients (ICC) and the coefficient of variance (COVL). The RFs’ correlation with volume and SUVmax was analyzed by calculating Pearson’s correlation coefficients. Results: DSC mean value was 0.75 ± 0.11 (0.45–0.92) between SAEB and operators and 0.78 ± 0.09 (0.36–0.97), among the four manual segmentations. The study showed high robustness (ICC > 0.9): (a) in 64.7% of RFs for segmentation methods using AR60, improved by applying SUVmax threshold of 40% (86.5%); (b) in 50.9% of RFs for different SUVmax thresholds using AR60; and (c) in 37% of RFs for discretization settings using different segmentation methods. Several RFs were not correlated with volume and SUVmax. Conclusions: RFs robustness to manual segmentation resulted higher in NET 68Ga-DOTA-TOC images compared to 18F-FDG PET/CT images. Forty percent SUVmax thresholds yield superior RFs stability among operators, however leading to a possible loss of biological information. SAEB segmentation appears to be an optimal alternative to manual segmentation, but further validations are needed. Finally, discretization settings highly impacted on RFs robustness and should always be stated.

AB - Objective: To identify the impact of segmentation methods and intensity discretization on radiomic features (RFs) extraction from 68Ga-DOTA-TOC PET images in patients with neuroendocrine tumors. Methods: Forty-nine patients were retrospectively analyzed. Tumor contouring was performed manually by four different operators and with a semi-automatic edge-based segmentation (SAEB) algorithm. Three SUVmax fixed thresholds (20, 30, 40%) were applied. Fifty-one RFs were extracted applying two different intensity rescale factors for gray-level discretization: one absolute (AR60 = SUV from 0 to 60) and one relative (RR = min-max of the VOI SUV). Dice similarity coefficient (DSC) was calculated to quantify segmentation agreement between different segmentation methods. The impact of segmentation and discretization on RFs was assessed by intra-class correlation coefficients (ICC) and the coefficient of variance (COVL). The RFs’ correlation with volume and SUVmax was analyzed by calculating Pearson’s correlation coefficients. Results: DSC mean value was 0.75 ± 0.11 (0.45–0.92) between SAEB and operators and 0.78 ± 0.09 (0.36–0.97), among the four manual segmentations. The study showed high robustness (ICC > 0.9): (a) in 64.7% of RFs for segmentation methods using AR60, improved by applying SUVmax threshold of 40% (86.5%); (b) in 50.9% of RFs for different SUVmax thresholds using AR60; and (c) in 37% of RFs for discretization settings using different segmentation methods. Several RFs were not correlated with volume and SUVmax. Conclusions: RFs robustness to manual segmentation resulted higher in NET 68Ga-DOTA-TOC images compared to 18F-FDG PET/CT images. Forty percent SUVmax thresholds yield superior RFs stability among operators, however leading to a possible loss of biological information. SAEB segmentation appears to be an optimal alternative to manual segmentation, but further validations are needed. Finally, discretization settings highly impacted on RFs robustness and should always be stated.

KW - Ga-DOTATOC PET/CT

KW - Neuroendocrine tumor

KW - Radiomics

KW - Robustness

KW - Semi-automatic segmentation

KW - Texture analysis

UR - http://www.scopus.com/inward/record.url?scp=85101880625&partnerID=8YFLogxK

U2 - 10.1186/s40658-021-00367-6

DO - 10.1186/s40658-021-00367-6

M3 - Article

AN - SCOPUS:85101880625

SN - 2197-7364

VL - 8

JO - EJNMMI physics

JF - EJNMMI physics

IS - 1

M1 - 21

ER -