A preliminary PET radiomics study of brain metastases using a fully automatic segmentation method.
Active contour
Biological target volume
Cancer
Positron emission tomography
Radiomics
Journal
BMC bioinformatics
ISSN: 1471-2105
Titre abrégé: BMC Bioinformatics
Pays: England
ID NLM: 100965194
Informations de publication
Date de publication:
16 Sep 2020
16 Sep 2020
Historique:
received:
18
06
2020
accepted:
09
07
2020
entrez:
17
9
2020
pubmed:
18
9
2020
medline:
22
10
2020
Statut:
epublish
Résumé
Positron Emission Tomography (PET) is increasingly utilized in radiomics studies for treatment evaluation purposes. Nevertheless, lesion volume identification in PET images is a critical and still challenging step in the process of radiomics, due to the low spatial resolution and high noise level of PET images. Currently, the biological target volume (BTV) is manually contoured by nuclear physicians, with a time expensive and operator-dependent procedure. This study aims to obtain BTVs from cerebral metastases in patients who underwent L-[ For predictive evaluation, 3 features (asphericity, low-intensity run emphasis, and complexity) were able to discriminate between responder and non-responder patients, after feature reduction and selection. Best performance in patient discrimination was obtained using the combination of the three selected features (sensitivity 81.23%, specificity 73.97%, and accuracy 78.27%) compared to the use of all features. Secondly, for follow-up evaluation, 8 features (SUV The proposed system is able i) to extract 108 features for each automatically segmented lesion and ii) to select a sub-panel of 11C-MET PET features (3 and 8 in the case of predictive and follow-up evaluation), with valuable association with patient outcome. We believe that our model can be useful to improve treatment response and prognosis evaluation, potentially allowing the personalization of cancer treatment plans.
Sections du résumé
BACKGROUND
BACKGROUND
Positron Emission Tomography (PET) is increasingly utilized in radiomics studies for treatment evaluation purposes. Nevertheless, lesion volume identification in PET images is a critical and still challenging step in the process of radiomics, due to the low spatial resolution and high noise level of PET images. Currently, the biological target volume (BTV) is manually contoured by nuclear physicians, with a time expensive and operator-dependent procedure. This study aims to obtain BTVs from cerebral metastases in patients who underwent L-[
RESULTS
RESULTS
For predictive evaluation, 3 features (asphericity, low-intensity run emphasis, and complexity) were able to discriminate between responder and non-responder patients, after feature reduction and selection. Best performance in patient discrimination was obtained using the combination of the three selected features (sensitivity 81.23%, specificity 73.97%, and accuracy 78.27%) compared to the use of all features. Secondly, for follow-up evaluation, 8 features (SUV
CONCLUSIONS
CONCLUSIONS
The proposed system is able i) to extract 108 features for each automatically segmented lesion and ii) to select a sub-panel of 11C-MET PET features (3 and 8 in the case of predictive and follow-up evaluation), with valuable association with patient outcome. We believe that our model can be useful to improve treatment response and prognosis evaluation, potentially allowing the personalization of cancer treatment plans.
Identifiants
pubmed: 32938360
doi: 10.1186/s12859-020-03647-7
pii: 10.1186/s12859-020-03647-7
pmc: PMC7493376
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
325Subventions
Organisme : Italian Ministry of Economic Development
ID : Grant No. F/090012/01-02/X36
Références
Comput Biol Med. 2014 Jul;50:76-96
pubmed: 24845019
Comput Biol Med. 2016 Feb 1;69:203-12
pubmed: 26773941
Nat Rev Mol Cell Biol. 2010 Mar;11(3):220-8
pubmed: 20177397
Eur J Nucl Med Mol Imaging. 2010 Mar;37(3):635-44
pubmed: 19882152
Phys Med Biol. 2017 Jul 7;62(13):5383-5402
pubmed: 28604372
Comput Biol Med. 2019 May;108:354-370
pubmed: 31054502
Int J Radiat Oncol Biol Phys. 2005 Sep 1;63(1):64-74
pubmed: 16111573
Tumori. 2020 Feb;106(1):33-38
pubmed: 31446858
Front Oncol. 2017 Nov 01;7:257
pubmed: 29164057
J Neurooncol. 2000 Sep;49(2):157-63
pubmed: 11206011
Eur J Nucl Med Mol Imaging. 2020 Jun;47(6):1400-1411
pubmed: 31773234
Radiology. 2016 Feb;278(2):563-77
pubmed: 26579733
Med Image Anal. 2018 Feb;44:177-195
pubmed: 29268169
Oncology. 2017;92(1):39-47
pubmed: 27832654
EJNMMI Res. 2018 Aug 22;8(1):86
pubmed: 30136163
Int J Radiat Oncol Biol Phys. 2018 Nov 15;102(4):1083-1089
pubmed: 29395627
Comput Biol Med. 2018 Nov 1;102:1-15
pubmed: 30219733
Surgeon. 2014 Jun;12(3):148-57
pubmed: 24629841
J Neurosurg. 2005 Sep;103(3):498-507
pubmed: 16235683
Clin Cancer Res. 2002 Apr;8(4):971-9
pubmed: 11948102
EBioMedicine. 2019 Aug;46:160-169
pubmed: 31395503
Eur Radiol Exp. 2019 Aug 7;3(1):35
pubmed: 31392526
J Nucl Med. 2009 May;50 Suppl 1:122S-50S
pubmed: 19403881
Int J Radiat Oncol Biol Phys. 2005 Oct 1;63(2):511-9
pubmed: 16168843
Artif Intell Med. 2019 Mar;94:67-78
pubmed: 30871684
J Neurosurg. 2002 Dec;97(5 Suppl):542-50
pubmed: 12507094
Comput Biol Med. 2020 May;120:103701
pubmed: 32217282
Biomed Res Int. 2014;2014:248505
pubmed: 24757667
Neuro Oncol. 2016 Mar;18(3):426-34
pubmed: 26243791
Br J Radiol. 2014 Jun;87(1038):20140065
pubmed: 24597512
J Nucl Med. 2017 Mar;58(3):365-366
pubmed: 27811126
Br J Cancer. 2010 Oct 12;103(8):1139-43
pubmed: 20877357
J Nucl Med. 2007 Jun;48(6):932-45
pubmed: 17504879
Cancer. 1997 Jan 1;79(1):115-26
pubmed: 8988735
Radiother Oncol. 2009 Jun;91(3):455-60
pubmed: 19339069
BMC Cancer. 2012 Jan 25;12:39
pubmed: 22277092
Indian J Nucl Med. 2011 Oct;26(4):171-6
pubmed: 23559710
Clin Radiol. 2017 Jan;72(1):3-10
pubmed: 27742105
Sci Rep. 2019 Oct 30;9(1):15666
pubmed: 31666650
Int J Radiat Oncol Biol Phys. 1998 Jul 15;41(5):989-95
pubmed: 9719107
Med Phys. 2017 Aug;44(8):4098-4111
pubmed: 28474819