Evaluation of a nnU-Net type automated clinical volumetric tumor segmentation tool for diffuse low-grade glioma follow-up.
Deep-learning
Follow-up
Glioma
Segmentation
Journal
Journal of neuroradiology = Journal de neuroradiologie
ISSN: 0150-9861
Titre abrégé: J Neuroradiol
Pays: France
ID NLM: 7705086
Informations de publication
Date de publication:
10 Jun 2023
10 Jun 2023
Historique:
received:
12
12
2022
revised:
30
05
2023
accepted:
30
05
2023
medline:
13
6
2023
pubmed:
13
6
2023
entrez:
12
6
2023
Statut:
aheadofprint
Résumé
Diffuse low-grade gliomas (DLGG) are characterized by a slow and continuous growth and always evolve towards an aggressive grade. Accurate prediction of the malignant transformation is essential as it requires immediate therapeutic intervention. One of its most precise predictors is the velocity of diameter expansion (VDE). Currently, the VDE is estimated either by linear measurements or by manual delineation of the DLGG on T2 FLAIR acquisitions. However, because of the DLGG's infiltrative nature and its blurred contours, manual measures are challenging and variable, even for experts. Therefore we propose an automated segmentation algorithm using a 2D nnU-Net, to 1) gain time and 2) standardize VDE assessment. The 2D nnU-Net was trained on 318 acquisitions (T2 FLAIR & 3DT1 longitudinal follow-up of 30 patients, including pre- & post-surgery acquisitions, different scanners, vendors, imaging parameters…). Automated vs. manual segmentation performance was evaluated on 167 acquisitions, and its clinical interest was validated by quantifying the amount of manual correction required after automated segmentation of 98 novel acquisitions. Automated segmentation showed a good performance with a mean Dice Similarity Coefficient (DSC) of 0.82±0.13 with manual segmentation and a substantial concordance between VDE calculations. Major manual corrections (i.e., DSC<0.7) were necessary only in 3/98 cases and 81% of the cases had a DSC>0.9. The proposed automated segmentation algorithm can successfully segment DLGG on highly variable MRI data. Although manual corrections are sometimes necessary, it provides a reliable, standardized and time-winning support for VDE extraction to asses DLGG growth.
Sections du résumé
BACKGROUND AND PURPOSE
OBJECTIVE
Diffuse low-grade gliomas (DLGG) are characterized by a slow and continuous growth and always evolve towards an aggressive grade. Accurate prediction of the malignant transformation is essential as it requires immediate therapeutic intervention. One of its most precise predictors is the velocity of diameter expansion (VDE). Currently, the VDE is estimated either by linear measurements or by manual delineation of the DLGG on T2 FLAIR acquisitions. However, because of the DLGG's infiltrative nature and its blurred contours, manual measures are challenging and variable, even for experts. Therefore we propose an automated segmentation algorithm using a 2D nnU-Net, to 1) gain time and 2) standardize VDE assessment.
MATERIALS AND METHODS
METHODS
The 2D nnU-Net was trained on 318 acquisitions (T2 FLAIR & 3DT1 longitudinal follow-up of 30 patients, including pre- & post-surgery acquisitions, different scanners, vendors, imaging parameters…). Automated vs. manual segmentation performance was evaluated on 167 acquisitions, and its clinical interest was validated by quantifying the amount of manual correction required after automated segmentation of 98 novel acquisitions.
RESULTS
RESULTS
Automated segmentation showed a good performance with a mean Dice Similarity Coefficient (DSC) of 0.82±0.13 with manual segmentation and a substantial concordance between VDE calculations. Major manual corrections (i.e., DSC<0.7) were necessary only in 3/98 cases and 81% of the cases had a DSC>0.9.
CONCLUSION
CONCLUSIONS
The proposed automated segmentation algorithm can successfully segment DLGG on highly variable MRI data. Although manual corrections are sometimes necessary, it provides a reliable, standardized and time-winning support for VDE extraction to asses DLGG growth.
Identifiants
pubmed: 37308338
pii: S0150-9861(23)00213-4
doi: 10.1016/j.neurad.2023.05.008
pii:
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Informations de copyright
Copyright © 2023 Elsevier Masson SAS. All rights reserved.
Déclaration de conflit d'intérêts
Declaration of Competing Interest Thomas Troalen is part of Siemens healthcare (France). Bénédicte Maréchal and Till Huelnhagen are part of the Advanced Clinical Imaging Technology group of Siemens Healthcare in Lausanne (Switzerland).