Implication of tumor morphology and MRI characteristics on the accuracy of automated versus human segmentation of GBM areas

Abstract An assessment scheme is proposed to evaluate GBM gross tumor core and T2-FLAIR hyper-intensity segmentations on preoperative multicentric MR images as a function of tumor morphology and MRI characteristics. 74 gross tumor core and T2-FLAIR hyper-intensity BraTS-Toolkit and DeepBraTumIA auto...

Full description

Saved in:
Bibliographic Details
Main Authors: Valeria Cerina, Chiara Benedetta Rui, Andrea Di Cristofori, Davide Ferlito, Giorgio Carrabba, Carlo Giussani, Gianpaolo Basso, Elisabetta De Bernardi
Format: Article
Language:English
Published: Nature Portfolio 2025-01-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-85400-9
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1832594798894120960
author Valeria Cerina
Chiara Benedetta Rui
Andrea Di Cristofori
Davide Ferlito
Giorgio Carrabba
Carlo Giussani
Gianpaolo Basso
Elisabetta De Bernardi
author_facet Valeria Cerina
Chiara Benedetta Rui
Andrea Di Cristofori
Davide Ferlito
Giorgio Carrabba
Carlo Giussani
Gianpaolo Basso
Elisabetta De Bernardi
author_sort Valeria Cerina
collection DOAJ
description Abstract An assessment scheme is proposed to evaluate GBM gross tumor core and T2-FLAIR hyper-intensity segmentations on preoperative multicentric MR images as a function of tumor morphology and MRI characteristics. 74 gross tumor core and T2-FLAIR hyper-intensity BraTS-Toolkit and DeepBraTumIA automatic segmentations, and 42 gross tumor core neurosurgeon manual segmentations were accordingly evaluated. Brats-Toolkit and DeepBraTumIA generally provide accurate segmentations, particularly for the most common round-shaped or well-demarked tumors, where: (1) gross tumor segmentation correctly includes necrosis and contrast enhanced tumor in 100% and 97.06% of cases (vs. 73.68% for manual segmentation) and wrongly includes healthy or non-tumor related tissues in 2.94% and 20.59% of cases (vs. 10.53% for manual segmentations); (2) T2-FLAIR hyper-intensity segmentations completely includes edema in 88.24% of cases for both software. MR image quality has little impact on the segmentation performance on these tumors. Conversely, on less common tumors with more complex tissue distribution and infiltrative behavior, manual segmentation works better than BraTS-Toolkit and DeepBraTumIA, and image quality has a larger impact on automatic segmentation performance. BraTS-Toolkit and DeepBraTumIA gross tumor segmentation properly includes necrosis and contrast enhanced areas in 50% and 37.50% of cases (vs. 66.67% for manual segmentation), all corresponding to higher image quality; T2-FLAIR hyper-intensity segmentation wrongly includes necrosis and contrast enhanced areas in 37.50% and 50% of cases.
format Article
id doaj-art-eb353224896e48229d11c1fbb494f203
institution Kabale University
issn 2045-2322
language English
publishDate 2025-01-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-eb353224896e48229d11c1fbb494f2032025-01-19T12:23:04ZengNature PortfolioScientific Reports2045-23222025-01-0115111210.1038/s41598-025-85400-9Implication of tumor morphology and MRI characteristics on the accuracy of automated versus human segmentation of GBM areasValeria Cerina0Chiara Benedetta Rui1Andrea Di Cristofori2Davide Ferlito3Giorgio Carrabba4Carlo Giussani5Gianpaolo Basso6Elisabetta De Bernardi7PhD program in Neuroscience, School of Medicine and Surgery, University of Milano-BicoccaNeurosurgery, Fondazione IRCCS San Gerardo dei TintoriPhD program in Neuroscience, School of Medicine and Surgery, University of Milano-BicoccaNeurosurgery, Fondazione IRCCS San Gerardo dei TintoriNeurosurgery, Fondazione IRCCS San Gerardo dei TintoriNeurosurgery, Fondazione IRCCS San Gerardo dei TintoriCENTRO STUDI DIPARTIMENTALE GBM-BI-TRACE (GlioBlastoMa-BIcocca-TRAnslational-CEnter)CENTRO STUDI DIPARTIMENTALE GBM-BI-TRACE (GlioBlastoMa-BIcocca-TRAnslational-CEnter)Abstract An assessment scheme is proposed to evaluate GBM gross tumor core and T2-FLAIR hyper-intensity segmentations on preoperative multicentric MR images as a function of tumor morphology and MRI characteristics. 74 gross tumor core and T2-FLAIR hyper-intensity BraTS-Toolkit and DeepBraTumIA automatic segmentations, and 42 gross tumor core neurosurgeon manual segmentations were accordingly evaluated. Brats-Toolkit and DeepBraTumIA generally provide accurate segmentations, particularly for the most common round-shaped or well-demarked tumors, where: (1) gross tumor segmentation correctly includes necrosis and contrast enhanced tumor in 100% and 97.06% of cases (vs. 73.68% for manual segmentation) and wrongly includes healthy or non-tumor related tissues in 2.94% and 20.59% of cases (vs. 10.53% for manual segmentations); (2) T2-FLAIR hyper-intensity segmentations completely includes edema in 88.24% of cases for both software. MR image quality has little impact on the segmentation performance on these tumors. Conversely, on less common tumors with more complex tissue distribution and infiltrative behavior, manual segmentation works better than BraTS-Toolkit and DeepBraTumIA, and image quality has a larger impact on automatic segmentation performance. BraTS-Toolkit and DeepBraTumIA gross tumor segmentation properly includes necrosis and contrast enhanced areas in 50% and 37.50% of cases (vs. 66.67% for manual segmentation), all corresponding to higher image quality; T2-FLAIR hyper-intensity segmentation wrongly includes necrosis and contrast enhanced areas in 37.50% and 50% of cases.https://doi.org/10.1038/s41598-025-85400-9GlioblastomaAutomatic segmentationSurgical planningBraTS-ToolkitDeepBraTumIAManual segmentation
spellingShingle Valeria Cerina
Chiara Benedetta Rui
Andrea Di Cristofori
Davide Ferlito
Giorgio Carrabba
Carlo Giussani
Gianpaolo Basso
Elisabetta De Bernardi
Implication of tumor morphology and MRI characteristics on the accuracy of automated versus human segmentation of GBM areas
Scientific Reports
Glioblastoma
Automatic segmentation
Surgical planning
BraTS-Toolkit
DeepBraTumIA
Manual segmentation
title Implication of tumor morphology and MRI characteristics on the accuracy of automated versus human segmentation of GBM areas
title_full Implication of tumor morphology and MRI characteristics on the accuracy of automated versus human segmentation of GBM areas
title_fullStr Implication of tumor morphology and MRI characteristics on the accuracy of automated versus human segmentation of GBM areas
title_full_unstemmed Implication of tumor morphology and MRI characteristics on the accuracy of automated versus human segmentation of GBM areas
title_short Implication of tumor morphology and MRI characteristics on the accuracy of automated versus human segmentation of GBM areas
title_sort implication of tumor morphology and mri characteristics on the accuracy of automated versus human segmentation of gbm areas
topic Glioblastoma
Automatic segmentation
Surgical planning
BraTS-Toolkit
DeepBraTumIA
Manual segmentation
url https://doi.org/10.1038/s41598-025-85400-9
work_keys_str_mv AT valeriacerina implicationoftumormorphologyandmricharacteristicsontheaccuracyofautomatedversushumansegmentationofgbmareas
AT chiarabenedettarui implicationoftumormorphologyandmricharacteristicsontheaccuracyofautomatedversushumansegmentationofgbmareas
AT andreadicristofori implicationoftumormorphologyandmricharacteristicsontheaccuracyofautomatedversushumansegmentationofgbmareas
AT davideferlito implicationoftumormorphologyandmricharacteristicsontheaccuracyofautomatedversushumansegmentationofgbmareas
AT giorgiocarrabba implicationoftumormorphologyandmricharacteristicsontheaccuracyofautomatedversushumansegmentationofgbmareas
AT carlogiussani implicationoftumormorphologyandmricharacteristicsontheaccuracyofautomatedversushumansegmentationofgbmareas
AT gianpaolobasso implicationoftumormorphologyandmricharacteristicsontheaccuracyofautomatedversushumansegmentationofgbmareas
AT elisabettadebernardi implicationoftumormorphologyandmricharacteristicsontheaccuracyofautomatedversushumansegmentationofgbmareas