Impact of brain segmentation methods on regional metabolism quantification in
Artificial intelligence
Magnetic resonance imaging
Metabolism
Positron emission tomography
Journal
EJNMMI research
ISSN: 2191-219X
Titre abrégé: EJNMMI Res
Pays: Germany
ID NLM: 101560946
Informations de publication
Date de publication:
05 Sep 2023
05 Sep 2023
Historique:
received:
14
06
2023
accepted:
28
08
2023
medline:
5
9
2023
pubmed:
5
9
2023
entrez:
5
9
2023
Statut:
epublish
Résumé
Accurate analysis of quantitative PET data plays a crucial role in studying small, specific brain structures. The integration of PET and MRI through an integrated PET/MR system presents an opportunity to leverage the benefits of precisely aligned structural MRI and molecular PET images in both spatial and temporal dimensions. However, in many clinical workflows, PET studies are often performed without the aid of individually matched structural MRI scans, primarily for the sake of convenience in the data collection and brain segmentation possesses. Currently, two commonly employed segmentation strategies for brain PET analysis are distinguished: methods with or without MRI registration and methods employing either atlas-based or individual-based algorithms. Moreover, the development of artificial intelligence (AI)-assisted methods for predicting brain segmentation holds promise but requires further validation of their efficiency and accuracy for clinical applications. This study aims to compare and evaluate the correlations, consistencies, and differences among the above-mentioned brain segmentation strategies in quantification of brain metabolism in Strong correlations were observed among all methods (r = 0.932 to 0.999, P < 0.001). The variances attributable to subject and brain region were higher than those caused by segmentation methods (P < 0.001). However, intraclass correlation coefficient (ICC)s between methods with or without MRI registration ranged from 0.924 to 0.975, while ICCs between methods with atlas- or individual-based algorithms ranged from 0.741 to 0.879. Brain regions exhibiting significant standardized uptake values (SUV) differences due to segmentation methods were the basal ganglia nuclei (maximum to 11.50 ± 4.67%), and various cerebral cortexes in temporal and occipital regions (maximum to 18.03 ± 5.52%). The AI-based method demonstrated high correlation (r = 0.998 and 0.999, P < 0.001) and ICC (0.998 and 0.997) with FreeSurfer, substantially reducing the time from 8.13 h to 57 s on per subject. Different segmentation methods may have impact on the calculation of brain metabolism in basal ganglia nuclei and specific cerebral cortexes. The AI-based approach offers improved efficiency and is recommended for its enhanced performance.
Sections du résumé
BACKGROUND
BACKGROUND
Accurate analysis of quantitative PET data plays a crucial role in studying small, specific brain structures. The integration of PET and MRI through an integrated PET/MR system presents an opportunity to leverage the benefits of precisely aligned structural MRI and molecular PET images in both spatial and temporal dimensions. However, in many clinical workflows, PET studies are often performed without the aid of individually matched structural MRI scans, primarily for the sake of convenience in the data collection and brain segmentation possesses. Currently, two commonly employed segmentation strategies for brain PET analysis are distinguished: methods with or without MRI registration and methods employing either atlas-based or individual-based algorithms. Moreover, the development of artificial intelligence (AI)-assisted methods for predicting brain segmentation holds promise but requires further validation of their efficiency and accuracy for clinical applications. This study aims to compare and evaluate the correlations, consistencies, and differences among the above-mentioned brain segmentation strategies in quantification of brain metabolism in
RESULTS
RESULTS
Strong correlations were observed among all methods (r = 0.932 to 0.999, P < 0.001). The variances attributable to subject and brain region were higher than those caused by segmentation methods (P < 0.001). However, intraclass correlation coefficient (ICC)s between methods with or without MRI registration ranged from 0.924 to 0.975, while ICCs between methods with atlas- or individual-based algorithms ranged from 0.741 to 0.879. Brain regions exhibiting significant standardized uptake values (SUV) differences due to segmentation methods were the basal ganglia nuclei (maximum to 11.50 ± 4.67%), and various cerebral cortexes in temporal and occipital regions (maximum to 18.03 ± 5.52%). The AI-based method demonstrated high correlation (r = 0.998 and 0.999, P < 0.001) and ICC (0.998 and 0.997) with FreeSurfer, substantially reducing the time from 8.13 h to 57 s on per subject.
CONCLUSIONS
CONCLUSIONS
Different segmentation methods may have impact on the calculation of brain metabolism in basal ganglia nuclei and specific cerebral cortexes. The AI-based approach offers improved efficiency and is recommended for its enhanced performance.
Identifiants
pubmed: 37668814
doi: 10.1186/s13550-023-01028-8
pii: 10.1186/s13550-023-01028-8
pmc: PMC10480127
doi:
Types de publication
Journal Article
Langues
eng
Pagination
79Subventions
Organisme : Beijing Natural Science Foundation
ID : 7224334
Organisme : National Natural Science Foundation of China
ID : 82130058
Organisme : HuiZhi Ascent Project of Xuanwu Hospital
ID : HZ2021ZCLJ005
Informations de copyright
© 2023. Springer-Verlag GmbH Germany, part of Springer Nature.
Références
Cereb Cortex. 2016 Oct;26(10):4004-14
pubmed: 26334050
IEEE Trans Comput Imaging. 2020;6:518-528
pubmed: 32055649
Psychiatriki. 2015 Oct-Dec;25(4):282-94
pubmed: 26709994
Neurosci Biobehav Rev. 2019 Mar;98:29-46
pubmed: 30611798
Eur J Nucl Med Mol Imaging. 2021 Jun;48(6):1726-1735
pubmed: 33388972
Neuroimage. 2011 Aug 1;57(3):856-65
pubmed: 21640841
Hum Brain Mapp. 2018 Dec;39(12):5126-5144
pubmed: 30076750
Neuroimage. 2005 Jul 1;26(3):839-51
pubmed: 15955494
Eur J Nucl Med Mol Imaging. 2020 Jul;47(7):1668-1677
pubmed: 31691843
Eur J Paediatr Neurol. 2018 Mar;22(2):321-326
pubmed: 29396173
Nat Neurosci. 2015 Dec;18(12):1853-60
pubmed: 26551545
Front Neuroinform. 2019 Oct 17;13:67
pubmed: 31749693
Eur Radiol. 2019 Mar;29(3):1355-1364
pubmed: 30242503
Bipolar Disord. 2006 Feb;8(1):65-74
pubmed: 16411982
BMC Neurosci. 2019 Aug 2;20(1):39
pubmed: 31375091
Neuron. 2013 Feb 6;77(3):586-95
pubmed: 23395382
Mol Imaging Radionucl Ther. 2018 Feb 1;27(1):10-18
pubmed: 29393048
Psychol Med. 2019 Apr;49(5):754-763
pubmed: 29734953
Eur J Nucl Med Mol Imaging. 2022 May;49(6):1930-1938
pubmed: 34939175
Comput Methods Programs Biomed. 1986 Aug;23(1):57-62
pubmed: 3638187
Psychiatry Res. 2015 Sep 30;233(3):299-305
pubmed: 26211622
Neuroimage. 2006 Jul 1;31(3):968-80
pubmed: 16530430
J Nucl Med. 2014 May 12;55(Supplement 2):2S-10S
pubmed: 24819419
Neuroimage. 2004;23 Suppl 1:S208-19
pubmed: 15501092
Hum Brain Mapp. 2007 Nov;28(11):1194-205
pubmed: 17266101
Psychiatry Res. 2015 Feb 28;231(2):176-83
pubmed: 25595222
Front Comput Neurosci. 2020 Feb 14;14:9
pubmed: 32116623
PLoS One. 2016 Oct 31;11(10):e0165719
pubmed: 27798694
Eur Radiol. 2021 Sep;31(9):7003-7011
pubmed: 33686474
Eur J Nucl Med Mol Imaging. 2020 Jun;47(6):1458-1467
pubmed: 31919633
Quant Imaging Med Surg. 2015 Apr;5(2):188-203
pubmed: 25853079
Eur J Nucl Med Mol Imaging. 2020 Sep;47(10):2440-2452
pubmed: 32157432
Psychiatry Investig. 2021 Jan;18(1):69-79
pubmed: 33561931
Eur J Nucl Med Mol Imaging. 2009 Mar;36 Suppl 1:S105-12
pubmed: 19104801
Cytometry A. 2010 Aug;77(8):733-42
pubmed: 20653013
Sci Data. 2014 Dec 09;1:140049
pubmed: 25977800
Neuroimage. 2012 Aug 15;62(2):774-81
pubmed: 22248573