Tracked 3D ultrasound and deep neural network-based thyroid segmentation reduce interobserver variability in thyroid volumetry.
Journal
PloS one
ISSN: 1932-6203
Titre abrégé: PLoS One
Pays: United States
ID NLM: 101285081
Informations de publication
Date de publication:
2022
2022
Historique:
received:
08
02
2022
accepted:
02
05
2022
entrez:
29
7
2022
pubmed:
30
7
2022
medline:
3
8
2022
Statut:
epublish
Résumé
Thyroid volumetry is crucial in the diagnosis, treatment, and monitoring of thyroid diseases. However, conventional thyroid volumetry with 2D ultrasound is highly operator-dependent. This study compares 2D and tracked 3D ultrasound with an automatic thyroid segmentation based on a deep neural network regarding inter- and intraobserver variability, time, and accuracy. Volume reference was MRI. 28 healthy volunteers (24-50 a) were scanned with 2D and 3D ultrasound (and by MRI) by three physicians (MD 1, 2, 3) with different experience levels (6, 4, and 1 a). In the 2D scans, the thyroid lobe volumes were calculated with the ellipsoid formula. A convolutional deep neural network (CNN) automatically segmented the 3D thyroid lobes. 26, 6, and 6 random lobe scans were used for training, validation, and testing, respectively. On MRI (T1 VIBE sequence) the thyroid was manually segmented by an experienced MD. MRI thyroid volumes ranged from 2.8 to 16.7ml (mean 7.4, SD 3.05). The CNN was trained to obtain an average Dice score of 0.94. The interobserver variability comparing two MDs showed mean differences for 2D and 3D respectively of 0.58 to 0.52ml (MD1 vs. 2), -1.33 to -0.17ml (MD1 vs. 3) and -1.89 to -0.70ml (MD2 vs. 3). Paired samples t-tests showed significant differences for 2D (p = .140, p = .002 and p = .002) and none for 3D (p = .176, p = .722 and p = .057). Intraobsever variability was similar for 2D and 3D ultrasound. Comparison of ultrasound volumes and MRI volumes showed a significant difference for the 2D volumetry of all MDs (p = .002, p = .009, p <.001), and no significant difference for 3D ultrasound (p = .292, p = .686, p = 0.091). Acquisition time was significantly shorter for 3D ultrasound. Tracked 3D ultrasound combined with a CNN segmentation significantly reduces interobserver variability in thyroid volumetry and increases the accuracy of the measurements with shorter acquisition times.
Identifiants
pubmed: 35905038
doi: 10.1371/journal.pone.0268550
pii: PONE-D-22-03907
pmc: PMC9337648
doi:
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
e0268550Déclaration de conflit d'intérêts
The authors have declared that no competing interests exist.
Références
Lancet. 1986 Feb 8;1(8476):307-10
pubmed: 2868172
IEEE J Biomed Health Inform. 2019 May;23(3):1110-1118
pubmed: 30113902
Thyroid. 2006 Jan;16(1):41-6
pubmed: 16487012
Int J Comput Assist Radiol Surg. 2017 Jun;12(6):1003-1011
pubmed: 28321804
Rev Med Chir Soc Med Nat Iasi. 2007 Oct-Dec;111(4):864-71
pubmed: 18389771
Thyroid. 2002 Aug;12(8):713-7
pubmed: 12225640
Dtsch Med Wochenschr. 1981 Oct 9;106(41):1338-40
pubmed: 7274082
Endocrine. 2016 Dec;54(3):751-756
pubmed: 27553050
Korean J Radiol. 2015 May-Jun;16(3):560-5
pubmed: 25995685
J Ultrasound Med. 2008 Dec;27(12):1727-33
pubmed: 19022998
Nuklearmedizin. 2007;46(1):1-7
pubmed: 17299648
Nuklearmedizin. 2014;53(3):99-104
pubmed: 24276677
Ultrasound Med Biol. 2014 Jul;40(7):1453-62
pubmed: 24768486
Neuroimage. 2019 Feb 1;186:713-727
pubmed: 30502445
J Nucl Med. 2021 Mar;62(3):304-312
pubmed: 33008929
J Ultrasound Med. 2004 Feb;23(2):247-54
pubmed: 14992362
Thyroid. 2001 Jun;11(6):569-74
pubmed: 11442004
Annu Int Conf IEEE Eng Med Biol Soc. 2019 Jul;2019:5832-5835
pubmed: 31947178
J Healthc Eng. 2018 Sep 23;2018:8087624
pubmed: 30344990
J Ultrasound Med. 2018 Jan;37(1):173-178
pubmed: 28736947
Thyroid. 2004 Feb;14(2):113-20
pubmed: 15068625
Nuklearmedizin. 2016 Dec 6;55(6):213-220
pubmed: 27922153
Annu Int Conf IEEE Eng Med Biol Soc. 2019 Jul;2019:5828-5831
pubmed: 31947177
IEEE Access. 2021;9:5119-5127
pubmed: 33747681
PLoS One. 2019 Jan 29;14(1):e0211215
pubmed: 30695052
IEEE Trans Biomed Eng. 2010 Jun;57(6):1348-57
pubmed: 20172782