{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,7]],"date-time":"2026-11-07T20:58:57Z","timestamp":1794085137737,"version":"4.8.0"},"reference-count":28,"publisher":"Public Library of Science (PLoS)","issue":"7","license":[{"start":{"date-parts":[[2022,7,29]],"date-time":"2022-07-29T00:00:00Z","timestamp":1659052800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/100010669","name":"H2020 LEIT Information and Communication Technologies","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100010669","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS ONE"],"abstract":"<jats:p>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\u201450 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 16.7<jats:italic>ml<\/jats:italic> (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 0.52<jats:italic>ml<\/jats:italic> (MD1 vs. 2), \u22121.33 to \u22120.17<jats:italic>ml<\/jats:italic> (MD1 vs. 3) and \u22121.89 to \u22120.70<jats:italic>ml<\/jats:italic> (MD2 vs. 3). Paired samples t-tests showed significant differences for 2D (<jats:italic>p<\/jats:italic> = .140, <jats:italic>p<\/jats:italic> = .002 and <jats:italic>p<\/jats:italic> = .002) and none for 3D (<jats:italic>p<\/jats:italic> = .176, <jats:italic>p<\/jats:italic> = .722 and <jats:italic>p<\/jats:italic> = .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 (<jats:italic>p<\/jats:italic> = .002, <jats:italic>p<\/jats:italic> = .009, <jats:italic>p<\/jats:italic> &lt;.001), and no significant difference for 3D ultrasound (<jats:italic>p<\/jats:italic> = .292, <jats:italic>p<\/jats:italic> = .686, <jats:italic>p<\/jats:italic> = 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.<\/jats:p>","DOI":"10.1371\/journal.pone.0268550","type":"journal-article","created":{"date-parts":[[2022,7,29]],"date-time":"2022-07-29T17:26:37Z","timestamp":1659115597000},"page":"e0268550","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":49,"title":["Tracked 3D ultrasound and deep neural network-based thyroid segmentation reduce interobserver variability in thyroid volumetry"],"prefix":"10.1371","volume":"17","author":[{"given":"Markus","family":"Kr\u00f6nke","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2895-0748","authenticated-orcid":true,"given":"Christine","family":"Eilers","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Desislava","family":"Dimova","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Melanie","family":"K\u00f6hler","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Gabriel","family":"Buschner","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lilit","family":"Schweiger","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7838-2967","authenticated-orcid":true,"given":"Lemonia","family":"Konstantinidou","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Marcus","family":"Makowski","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"James","family":"Nagarajah","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nassir","family":"Navab","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wolfgang","family":"Weber","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0915-0510","authenticated-orcid":true,"given":"Thomas","family":"Wendler","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2022,7,29]]},"reference":[{"key":"pone.0268550.ref001","doi-asserted-by":"crossref","first-page":"751","DOI":"10.1007\/s12020-016-1074-7","article-title":"Calculation of therapeutic activity of radioiodine in Graves\u2019 disease by means of Marinelli\u2019s formula, using technetium (99mTc) scintigraphy","volume":"54","author":"P. Szumowski","year":"2016","journal-title":"Endocrine"},{"key":"pone.0268550.ref002","first-page":"1338","article-title":"Volumetric analysis of thyroid lobes by real-time ultrasound (author\u2019s transl)","volume":"106","author":"J. Brunn","year":"1981","journal-title":"Dtsch Med Wochenschr"},{"key":"pone.0268550.ref003","doi-asserted-by":"crossref","first-page":"213","DOI":"10.3413\/Nukmed-0823-16-04","article-title":"Radioiodtherapie bei benignen Schilddr\u00fcsenerkrankungen (Version 5)*: DGN-Handlungsempfehlung (S1-Leitlinie)","volume":"55","author":"M. Dietlein","year":"2016","journal-title":"Nuklearmedizin"},{"key":"pone.0268550.ref004","doi-asserted-by":"crossref","first-page":"304","DOI":"10.2967\/jnumed.120.243170","article-title":"The Role of Nuclear Medicine in the Clinical Management of Benign Thyroid Disorders, Part 1: Hyperthyroidism","volume":"62","author":"G. Mariani","year":"2021","journal-title":"J Nucl Med"},{"key":"pone.0268550.ref005","doi-asserted-by":"crossref","first-page":"247","DOI":"10.7863\/jum.2004.23.2.247","article-title":"Three-dimensional ultrasonography for volume measurement of thyroid nodules in children","volume":"23","author":"A. Lyshchik","year":"2004","journal-title":"J Ultrasound Med"},{"key":"pone.0268550.ref006","doi-asserted-by":"crossref","first-page":"1003","DOI":"10.1007\/s11548-017-1561-1","article-title":"On the reproducibility of expert-operated and robotic ultrasound acquisitions","volume":"12","author":"R. Kojcev","year":"2017","journal-title":"Int J Comput Assist Radiol Surg"},{"key":"pone.0268550.ref007","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1055\/s-0037-1616621","article-title":"Intra- and interobserver variability of thyroid volume measurements in healthy adults by 2D versus 3D ultrasound","volume":"46","author":"P. Andermann","year":"2007","journal-title":"Nuklearmedizin. Nuclear Medicine"},{"key":"pone.0268550.ref008","doi-asserted-by":"crossref","first-page":"1348","DOI":"10.1109\/TBME.2010.2041003","article-title":"Thyroid Segmentation and Volume Estimation in Ultrasound Images","volume":"57","author":"C. Chang","year":"2010","journal-title":"IEEE Transactions On Biomedical Engineering"},{"key":"pone.0268550.ref009","article-title":"Evaluation of Commonly Used Algorithms for Thyroid Ultrasound Images Segmentation and Improvement Using Machine Learning Approaches","author":"P. Poudel","year":"2018","journal-title":"Journal Of Healthcare Engineering"},{"key":"pone.0268550.ref010","first-page":"1","article-title":"Illanes Parametrical modelling for texture characterization\u2014A novel approach applied to ultrasound thyroid segmentation","volume":"14","year":"2019","journal-title":"PLOS ONE"},{"key":"pone.0268550.ref011","doi-asserted-by":"crossref","first-page":"1110","DOI":"10.1109\/JBHI.2018.2864896","article-title":"Anatomical Structure Segmentation in Ultrasound Volumes Using Cross Frame Belief Propagating Iterative Random Walks","volume":"23","author":"D. China","year":"2019","journal-title":"IEEE Journal Of Biomedical And Health Informatics"},{"key":"pone.0268550.ref012","doi-asserted-by":"crossref","unstructured":"Mahmoodian, N., Poudel, P., Illanes, A. & Friebe, M. Higher Order Statistical Analysis for Thyroid Texture Classification and Segmentation in 2D ultrasound Images. 2019 41st Annual International Conference Of The IEEE Engineering In Medicine And Biology Society (EMBC). pp. 5832-5835 (2019)","DOI":"10.1109\/EMBC.2019.8857380"},{"key":"pone.0268550.ref013","doi-asserted-by":"crossref","unstructured":"Poudel, P., Illanes, A., Sadeghi, M. & Friebe, M. Patch Based Texture Classification of Thyroid Ultrasound Images using Convolutional Neural Network. 2019 41st Annual International Conference Of The IEEE Engineering In Medicine And Biology Society (EMBC). pp. 5828-5831 (2019)","DOI":"10.1109\/EMBC.2019.8857929"},{"key":"pone.0268550.ref014","doi-asserted-by":"crossref","first-page":"5119","DOI":"10.1109\/ACCESS.2020.3045906","article-title":"Automatic Deep Learning Semantic Segmentation of Ultrasound Thyroid Cineclips Using Recurrent Fully Convolutional Networks","volume":"9","author":"J. Webb","year":"2021","journal-title":"IEEE Access"},{"key":"pone.0268550.ref015","doi-asserted-by":"crossref","first-page":"713","DOI":"10.1089\/105072502760258695","article-title":"Thyroid volume measurement in patients prior to radioiodine therapy: comparison between three-dimensional magnetic resonance imaging and ultrasonography","volume":"12","author":"P. Reinartz","year":"2002","journal-title":"Thyroid"},{"key":"pone.0268550.ref016","first-page":"864","article-title":"Thyroid ultrasonography. Interobserver and intraobserver variations","volume":"111","author":"C. Vulpoi","year":"2007","journal-title":"Rev Med Chir Soc Med Nat Iasi"},{"key":"pone.0268550.ref017","doi-asserted-by":"crossref","first-page":"173","DOI":"10.1002\/jum.14316","article-title":"Intraobserver and Interobserver Variability in Ultrasound Measurements of Thyroid Nodules","volume":"37","author":"H. Lee","year":"2018","journal-title":"J Ultrasound Med"},{"key":"pone.0268550.ref018","doi-asserted-by":"crossref","first-page":"560","DOI":"10.3348\/kjr.2015.16.3.560","article-title":"Inter-observer variation in ultrasound measurement of the volume and diameter of thyroid nodules","volume":"16","author":"Y. Choi","year":"2015","journal-title":"Korean J Radiol"},{"key":"pone.0268550.ref019","doi-asserted-by":"crossref","first-page":"41","DOI":"10.1089\/thy.2006.16.41","article-title":"A novel thyroid phantom for ultrasound volumetry: determination of intraobserver and interobserver variability","volume":"16","author":"S. Schl\u00f6gl","year":"2006","journal-title":"Thyroid"},{"key":"pone.0268550.ref020","doi-asserted-by":"crossref","first-page":"113","DOI":"10.1089\/105072504322880346","article-title":"Accuracy of three-dimensional ultrasound for thyroid volume measurement in children and adolescents","volume":"14","author":"A. Lyshchik","year":"2004","journal-title":"Thyroid"},{"key":"pone.0268550.ref021","doi-asserted-by":"crossref","first-page":"1727","DOI":"10.7863\/jum.2008.27.12.1727","article-title":"Thyroid volumetric quantification: comparative evaluation between conventional and volumetric ultrasonography","volume":"27","author":"R. Malago","year":"2008","journal-title":"J Ultrasound Med"},{"key":"pone.0268550.ref022","doi-asserted-by":"crossref","first-page":"569","DOI":"10.1089\/105072501750302877","article-title":"The use of three-dimensional ultrasound for thyroid volumetry","volume":"11","author":"S. Schl\u00f6gl","year":"2001","journal-title":"Thyroid"},{"key":"pone.0268550.ref023","doi-asserted-by":"crossref","first-page":"1453","DOI":"10.1016\/j.ultrasmedbio.2014.02.013","article-title":"Multimodal evaluation of 2-D and 3-D ultrasound, computed tomography and magnetic resonance imaging in measurements of the thyroid volume using universally applicable cross-sectional imaging software: a phantom study","volume":"40","author":"M. Freesmeyer","year":"2014","journal-title":"Ultrasound Med Biol"},{"key":"pone.0268550.ref024","doi-asserted-by":"crossref","first-page":"713","DOI":"10.1016\/j.neuroimage.2018.11.042","article-title":"Alzheimer\u2019s Disease Neuroimaging Initiative QuickNAT: A fully convolutional network for quick and accurate segmentation of neuroanatomy","volume":"186","author":"A. Guha Roy","year":"2019","journal-title":"Neuroimage"},{"key":"pone.0268550.ref025","doi-asserted-by":"crossref","unstructured":"Ronneberger, O., Fischer, P. & Brox, T. U-Net: Convolutional Networks for Biomedical Image Segmentation. Medical Image Computing And Computer-Assisted Intervention\u2014MICCAI 2015. pp. 234-241 (2015)","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"pone.0268550.ref026","doi-asserted-by":"crossref","unstructured":"Castro, E., Cardoso, J. & Pereira, J. Elastic deformations for data augmentation in breast cancer mass detection. 2018 IEEE EMBS International Conference On Biomedical Health Informatics (BHI). pp. 230-234 (2018)","DOI":"10.1109\/BHI.2018.8333411"},{"key":"pone.0268550.ref027","doi-asserted-by":"crossref","first-page":"307","DOI":"10.1016\/S0140-6736(86)90837-8","article-title":"Statistical methods for assessing agreement between two methods of clinical measurement","volume":"1","author":"J. Bland","year":"1986","journal-title":"Lancet"},{"key":"pone.0268550.ref028","doi-asserted-by":"crossref","first-page":"99","DOI":"10.3413\/Nukmed-0615-13-08","article-title":"3D ultrasonography is as accurate as low-dose CT in thyroid volumetry","volume":"53","author":"K. Licht","year":"2014","journal-title":"Nuklearmedizin"}],"container-title":["PLOS ONE"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0268550","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,7,29]],"date-time":"2022-07-29T17:27:03Z","timestamp":1659115623000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0268550"}},"subtitle":[],"editor":[{"given":"Sathishkumar","family":"V E","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2022,7,29]]},"references-count":28,"journal-issue":{"issue":"7","published-online":{"date-parts":[[2022,7,29]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0268550","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,7,29]]}}}