TY - JOUR
T1 - Standardised lesion segmentation for imaging biomarker quantitation
T2 - a consensus recommendation from ESR and EORTC
AU - deSouza, Nandita M.
AU - van der Lugt, Aad
AU - Deroose, Christophe M.
AU - Alberich-Bayarri, Angel
AU - Bidaut, Luc
AU - Fournier, Laure
AU - Costaridou, Lena
AU - Oprea-Lager, Daniela E.
AU - Kotter, Elmar
AU - Smits, Marion
AU - Mayerhoefer, Marius E.
AU - Boellaard, Ronald
AU - Caroli, Anna
AU - de Geus-Oei, Lioe-Fee
AU - Kunz, Wolfgang G.
AU - Oei, Edwin H.
AU - Lecouvet, Frederic
AU - Franca, Manuela
AU - Loewe, Christian
AU - Lopci, Egesta
AU - Caramella, Caroline
AU - Persson, Anders
AU - Golay, Xavier
AU - Dewey, Marc
AU - O’Connor, James P. B.
AU - deGraaf, Pim
AU - European Society of Radiology
AU - Gatidis, Sergios
AU - European Organisation for Research and Treatment of Cancer
AU - Zahlmann, Gudrun
N1 - Funding Information: We are extremely grateful to ESR’s Jonathan Clark for his effort in coordinating this project and enabling the surveys for distribution. The work would not have been possible without his administrative assistance. Nandita M. deSouza, Aad van der Lugt, Angel Alberich-Bayarri, Laure Fournier, Lena Costaridou, Elmar Kotter, Marion Smits, Marius E. Mayerhoefer, Ronald Boellaard, Anna Caroli, Edwin H. Oei, Manuela Franca, Christian Loewe, Anders Persson, Xavier Golay, Marc Dewey, James P.B. O’Connor, Pim deGraaf, Sergios Gatidis, Gudrun Zahlmann. Nandita M. deSouza, Christophe M Deroose, Luc Bidaut, Laure Fournier, Daniela E. Oprea-Lager, Marion Smits, Lioe-Fee de Geus-Oei, Wolfgang G Kunz, Frederic Lecouvet, Egesta Lopci, Caroline Caramella. Nandita M. deSouza, Ronald Boellaard, Xavier Golay, Gudrun Zahlmann. The paper was endorsed by the ESR Executive Council in August 2022. Funding Information: The following authors acknowledge funding that allowed their participation in this work: AA-B acknowledges financial support from the European Union’s Horizon 2020 research and innovation programme under grant agreement: PRIMAGE (nº 826494), ProCancer-I (nº 952159), ChAImeleon (nº 952172), and the Horizon Europe framework programme 2021 under grant agreement: RadioVal (nº 101057699). MEM is funded in part through the NIH/NCI Cancer Center Support Grant P30 CA008748. XG is supported by the National Institute for Health Research University College London Hospitals Biomedical Research Centre. LF is funded in part by the French government under management of the Agence Nationale de la Recherche as part of the “Investissements d’avenir” program, reference ANR19-P3IA-0001 (PRAIRIE 3IA Institute). Publisher Copyright: © 2022, The Author(s).
PY - 2022/12/1
Y1 - 2022/12/1
N2 - Background: Lesion/tissue segmentation on digital medical images enables biomarker extraction, image-guided therapy delivery, treatment response measurement, and training/validation for developing artificial intelligence algorithms and workflows. To ensure data reproducibility, criteria for standardised segmentation are critical but currently unavailable. Methods: A modified Delphi process initiated by the European Imaging Biomarker Alliance (EIBALL) of the European Society of Radiology (ESR) and the European Organisation for Research and Treatment of Cancer (EORTC) Imaging Group was undertaken. Three multidisciplinary task forces addressed modality and image acquisition, segmentation methodology itself, and standards and logistics. Devised survey questions were fed via a facilitator to expert participants. The 58 respondents to Round 1 were invited to participate in Rounds 2–4. Subsequent rounds were informed by responses of previous rounds. Results/conclusions: Items with ≥ 75% consensus are considered a recommendation. These include system performance certification, thresholds for image signal-to-noise, contrast-to-noise and tumour-to-background ratios, spatial resolution, and artefact levels. Direct, iterative, and machine or deep learning reconstruction methods, use of a mixture of CE marked and verified research tools were agreed and use of specified reference standards and validation processes considered essential. Operator training and refreshment were considered mandatory for clinical trials and clinical research. Items with a 60–74% agreement require reporting (site-specific accreditation for clinical research, minimal pixel number within lesion segmented, use of post-reconstruction algorithms, operator training refreshment for clinical practice). Items with ≤ 60% agreement are outside current recommendations for segmentation (frequency of system performance tests, use of only CE-marked tools, board certification of operators, frequency of operator refresher training). Recommendations by anatomical area are also specified.
AB - Background: Lesion/tissue segmentation on digital medical images enables biomarker extraction, image-guided therapy delivery, treatment response measurement, and training/validation for developing artificial intelligence algorithms and workflows. To ensure data reproducibility, criteria for standardised segmentation are critical but currently unavailable. Methods: A modified Delphi process initiated by the European Imaging Biomarker Alliance (EIBALL) of the European Society of Radiology (ESR) and the European Organisation for Research and Treatment of Cancer (EORTC) Imaging Group was undertaken. Three multidisciplinary task forces addressed modality and image acquisition, segmentation methodology itself, and standards and logistics. Devised survey questions were fed via a facilitator to expert participants. The 58 respondents to Round 1 were invited to participate in Rounds 2–4. Subsequent rounds were informed by responses of previous rounds. Results/conclusions: Items with ≥ 75% consensus are considered a recommendation. These include system performance certification, thresholds for image signal-to-noise, contrast-to-noise and tumour-to-background ratios, spatial resolution, and artefact levels. Direct, iterative, and machine or deep learning reconstruction methods, use of a mixture of CE marked and verified research tools were agreed and use of specified reference standards and validation processes considered essential. Operator training and refreshment were considered mandatory for clinical trials and clinical research. Items with a 60–74% agreement require reporting (site-specific accreditation for clinical research, minimal pixel number within lesion segmented, use of post-reconstruction algorithms, operator training refreshment for clinical practice). Items with ≤ 60% agreement are outside current recommendations for segmentation (frequency of system performance tests, use of only CE-marked tools, board certification of operators, frequency of operator refresher training). Recommendations by anatomical area are also specified.
KW - Modality-specific
KW - Organ-specific
KW - Region of interest
KW - Segmentation and standardisation
KW - mDelphi
UR - http://www.scopus.com/inward/record.url?scp=85139386113&partnerID=8YFLogxK
U2 - https://doi.org/10.1186/s13244-022-01287-4
DO - https://doi.org/10.1186/s13244-022-01287-4
M3 - Article
C2 - 36194301
SN - 1869-4101
VL - 13
SP - 159
JO - Insights into imaging
JF - Insights into imaging
IS - 1
M1 - 159
ER -