{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,1]],"date-time":"2026-04-01T17:26:42Z","timestamp":1775064402807,"version":"3.50.1"},"reference-count":25,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2021,11,25]],"date-time":"2021-11-25T00:00:00Z","timestamp":1637798400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2021,11,25]],"date-time":"2021-11-25T00:00:00Z","timestamp":1637798400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"Sichuan Science and Technology Program of China","award":["2020JDRC0042"],"award-info":[{"award-number":["2020JDRC0042"]}]},{"name":"1.3.5 project for disciplines of excellence in West China Hospital","award":["ZYGD18016"],"award-info":[{"award-number":["ZYGD18016"]}]},{"name":"1.3.5 project for disciplines of excellence in West China Hospital","award":["2021HXFH033"],"award-info":[{"award-number":["2021HXFH033"]}]},{"DOI":"10.13039\/501100013076","name":"National Major Science and Technology Projects of China","doi-asserted-by":"crossref","award":["2018AAA0100201"],"award-info":[{"award-number":["2018AAA0100201"]}],"id":[{"id":"10.13039\/501100013076","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["BMC Med Imaging"],"published-print":{"date-parts":[[2021,12]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:sec><jats:title>Background<\/jats:title><jats:p><jats:sup>99m<\/jats:sup>Tc-pertechnetate thyroid scintigraphy is a valid complementary avenue for evaluating thyroid disease in the clinic, the image feature of thyroid scintigram is relatively simple but the interpretation still has a moderate consistency among physicians. Thus, we aimed to develop an artificial intelligence (AI) system to automatically classify the four patterns of thyroid scintigram.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>We collected 3087 thyroid scintigrams from center 1 to construct the training dataset (n\u2009=\u20092468) and internal validating dataset (n\u2009=\u2009619), and another 302 cases from center 2 as external validating datasets. Four pre-trained neural networks that included ResNet50, DenseNet169, InceptionV3, and InceptionResNetV2 were implemented to construct AI models. The models were trained separately with transfer learning. We evaluated each model\u2019s performance with metrics as following: accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), recall, precision, and F1-score.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>The overall accuracy of four pre-trained neural networks in classifying four common uptake patterns of thyroid scintigrams all exceeded 90%, and the InceptionV3 stands out from others. It reached the highest performance with an overall accuracy of 92.73% for internal validation and 87.75% for external validation, respectively. As for each category of thyroid scintigrams, the area under the receiver operator characteristic curve (AUC) was 0.986 for \u2018diffusely increased,\u2019 0.997 for \u2018diffusely decreased,\u2019 0.998 for \u2018focal increased,\u2019 and 0.945 for \u2018heterogeneous uptake\u2019 in internal validation, respectively. Accordingly, the corresponding performances also obtained an ideal result of 0.939, 1.000, 0.974, and 0.915 in external validation, respectively.<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusions<\/jats:title><jats:p>Deep convolutional neural network-based AI model represented considerable performance in the classification of thyroid scintigrams, which may help physicians improve the interpretation of thyroid scintigrams more consistently and efficiently.<\/jats:p><\/jats:sec>","DOI":"10.1186\/s12880-021-00710-4","type":"journal-article","created":{"date-parts":[[2021,11,25]],"date-time":"2021-11-25T15:03:11Z","timestamp":1637852591000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":15,"title":["Automatic differentiation of thyroid scintigram by deep convolutional neural network: a dual center study"],"prefix":"10.1186","volume":"21","author":[{"given":"Pei","family":"Yang","sequence":"first","affiliation":[]},{"given":"Yong","family":"Pi","sequence":"additional","affiliation":[]},{"given":"Tao","family":"He","sequence":"additional","affiliation":[]},{"given":"Jiangming","family":"Sun","sequence":"additional","affiliation":[]},{"given":"Jianan","family":"Wei","sequence":"additional","affiliation":[]},{"given":"Yongzhao","family":"Xiang","sequence":"additional","affiliation":[]},{"given":"Lisha","family":"Jiang","sequence":"additional","affiliation":[]},{"given":"Lin","family":"Li","sequence":"additional","affiliation":[]},{"given":"Zhang","family":"Yi","sequence":"additional","affiliation":[]},{"given":"Zhen","family":"Zhao","sequence":"additional","affiliation":[]},{"given":"Huawei","family":"Cai","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2021,11,25]]},"reference":[{"key":"710_CR1","doi-asserted-by":"publisher","first-page":"2514","DOI":"10.1007\/s00259-019-04472-8","volume":"46","author":"L Giovanella","year":"2019","unstructured":"Giovanella L, Avram AM, Iakovou I, et al. EANM practice guideline\/SNMMI procedure standard for RAIU and thyroid scintigraphy. Eur J Nucl Med Mol Imaging. 2019;46:2514\u201325.","journal-title":"Eur J Nucl Med Mol Imaging"},{"key":"710_CR2","doi-asserted-by":"publisher","first-page":"181","DOI":"10.1097\/01.rlu.0000114530.12565.5b","volume":"29","author":"JR Smith","year":"2004","unstructured":"Smith JR, Oates E. Radionuclide imaging of the thyroid gland: patterns, pearls, and pitfalls. Clin Nucl Med. 2004;29:181\u201393.","journal-title":"Clin Nucl Med"},{"key":"710_CR3","doi-asserted-by":"publisher","first-page":"1155","DOI":"10.1007\/s40618-017-0750-7","volume":"40","author":"KA Patel","year":"2017","unstructured":"Patel KA, Warren R, Brooke A, et al. Interpretation of thyroid scintigraphy is inconsistent among endocrinologists. J Endocrinol Invest. 2017;40:1155\u20137.","journal-title":"J Endocrinol Invest"},{"key":"710_CR4","doi-asserted-by":"publisher","first-page":"e0204596","DOI":"10.1371\/journal.pone.0204596","volume":"13","author":"M Dong","year":"2018","unstructured":"Dong M, Huang X, Xu B. Unsupervised speech recognition through spike-timing-dependent plasticity in a convolutional spiking neural network. PLoS ONE. 2018;13:e0204596.","journal-title":"PLoS ONE."},{"key":"710_CR5","doi-asserted-by":"publisher","first-page":"13080","DOI":"10.1038\/s41598-019-49411-7","volume":"9","author":"DA Frank","year":"2019","unstructured":"Frank DA, Chrysochou P, Mitkidis P, Ariely D. Human decision-making biases in the moral dilemmas of autonomous vehicles. Sci Rep. 2019;9:13080.","journal-title":"Sci Rep"},{"key":"710_CR6","doi-asserted-by":"publisher","first-page":"508","DOI":"10.1126\/science.aam6960","volume":"356","author":"M Moravcik","year":"2017","unstructured":"Moravcik M, Schmid M, Burch N, et al. DeepStack: Expert-level artificial intelligence in heads-up no-limit poker. Science. 2017;356:508\u201313.","journal-title":"Science"},{"key":"710_CR7","doi-asserted-by":"publisher","first-page":"1654","DOI":"10.1016\/j.jcmg.2018.01.020","volume":"11","author":"J Betancur","year":"2018","unstructured":"Betancur J, Commandeur F, Motlagh M, et al. Deep learning for prediction of obstructive disease from fast myocardial perfusion SPECT: a multicenter study. JACC Cardiovasc Imaging. 2018;11:1654\u201363.","journal-title":"JACC Cardiovasc Imaging"},{"key":"710_CR8","doi-asserted-by":"publisher","first-page":"549","DOI":"10.1093\/ehjci\/jez177","volume":"21","author":"LH Hu","year":"2020","unstructured":"Hu LH, Betancur J, Sharir T, et al. Machine learning predicts per-vessel early coronary revascularization after fast myocardial perfusion SPECT: results from multicentre REFINE SPECT registry. Eur Heart J Cardiovasc Imaging. 2020;21:549\u201359.","journal-title":"Eur Heart J Cardiovasc Imaging"},{"key":"710_CR9","doi-asserted-by":"publisher","first-page":"101784","DOI":"10.1016\/j.media.2020.101784","volume":"65","author":"Y Pi","year":"2020","unstructured":"Pi Y, Zhao Z, Xiang Y, Li Y, Cai H, Yi Z. Automated diagnosis of bone metastasis based on multi-view bone scans using attention-augmented deep neural networks. Med Image Anal. 2020;65:101784.","journal-title":"Med Image Anal."},{"key":"710_CR10","first-page":"6212759","volume":"2019","author":"L Ma","year":"2019","unstructured":"Ma L, Ma C, Liu Y, Wang X. Thyroid diagnosis from SPECT images using convolutional neural network with optimization. Comput Intell Neurosci. 2019;2019:6212759.","journal-title":"Comput Intell Neurosci"},{"key":"710_CR11","doi-asserted-by":"publisher","first-page":"1343","DOI":"10.1089\/thy.2016.0229","volume":"26","author":"DS Ross","year":"2016","unstructured":"Ross DS, Burch HB, Cooper DS, et al. 2016 American Thyroid Association Guidelines for diagnosis and management of hyperthyroidism and other causes of thyrotoxicosis. Thyroid. 2016;26:1343\u2013421.","journal-title":"Thyroid"},{"key":"710_CR12","first-page":"233","volume":"36","author":"F Ianni","year":"2013","unstructured":"Ianni F, Perotti G, Prete A, et al. Thyroid scintigraphy: an old tool is still the gold standard for an effective diagnosis of autonomously functioning thyroid nodules. J Endocrinol Invest. 2013;36:233\u20136.","journal-title":"J Endocrinol Invest"},{"key":"710_CR13","doi-asserted-by":"publisher","first-page":"857","DOI":"10.1148\/rg.234025716","volume":"23","author":"CM Intenzo","year":"2003","unstructured":"Intenzo CM, dePapp AE, Jabbour S, Miller JL, Kim SM, Capuzzi DM. Scintigraphic manifestations of thyrotoxicosis. Radiographics. 2003;23:857\u201369.","journal-title":"Radiographics"},{"key":"710_CR14","doi-asserted-by":"publisher","first-page":"1048","DOI":"10.1016\/j.mayocp.2018.10.011","volume":"94","author":"A Sharma","year":"2019","unstructured":"Sharma A, Stan MN. Thyrotoxicosis: diagnosis and management. Mayo Clin Proc. 2019;94:1048\u201364.","journal-title":"Mayo Clin Proc"},{"key":"710_CR15","unstructured":"Zhang H, Cisse M, Dauphin YN, Lopezpaz D. mixup: beyond empirical risk minimization. arXiv preprint arXiv:1710.09412, 2017."},{"key":"710_CR16","doi-asserted-by":"crossref","unstructured":"He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. Paper presented at: computer vision and pattern recognition. 2016.","DOI":"10.1109\/CVPR.2016.90"},{"key":"710_CR17","doi-asserted-by":"crossref","unstructured":"Huang G, Liu Z, Der Maaten LV, Weinberger KQ. Densely connected convolutional networks. Paper presented at: computer vision and pattern recognition. 2017.","DOI":"10.1109\/CVPR.2017.243"},{"key":"710_CR18","doi-asserted-by":"crossref","unstructured":"Szegedy C, Vanhoucke V, Ioffe S, Shlens J, Wojna Z. Rethinking the inception architecture for computer vision. Paper presented at: IEEE conference on computer vision and pattern recognition. 2016.","DOI":"10.1109\/CVPR.2016.308"},{"key":"710_CR19","doi-asserted-by":"crossref","unstructured":"Szegedy C, Ioffe S, Vanhoucke V, Alemi AA. Inception-v4, Inception-ResNet and the impact of residual connections on learning. Paper presented at: National conference on artificial intelligence. 2016.","DOI":"10.1609\/aaai.v31i1.11231"},{"key":"710_CR20","unstructured":"Kingma DP, Ba J. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980, 2014."},{"key":"710_CR21","first-page":"1929","volume":"15","author":"N Srivastava","year":"2014","unstructured":"Srivastava N, Hinton GE, Krizhevsky A, Sutskever I, Salakhutdinov RJ. Dropout: a simple way to prevent neural networks from overfitting. JMLR. 2014;15:1929\u201358.","journal-title":"JMLR"},{"key":"710_CR22","doi-asserted-by":"publisher","first-page":"459","DOI":"10.1016\/S0140-6736(03)14073-1","volume":"362","author":"DS Cooper","year":"2003","unstructured":"Cooper DS. Hyperthyroidism. Lancet. 2003;362:459\u201368.","journal-title":"Lancet"},{"key":"710_CR23","doi-asserted-by":"publisher","first-page":"657","DOI":"10.4158\/EP-2018-0544","volume":"25","author":"S Teelucksingh","year":"2019","unstructured":"Teelucksingh S, Motilal MS, Bailey H, et al. Management of thyrotoxicosis among general practitioners in trinidad compared with 2016 American Thyroid Association Guidelines for Hyperthyroidism. Endocr Pract. 2019;25:657\u201362.","journal-title":"Endocr Pract"},{"key":"710_CR24","doi-asserted-by":"publisher","first-page":"957","DOI":"10.1148\/radiographics.21.4.g01jl17957","volume":"21","author":"M Charles","year":"2001","unstructured":"Charles M, Intenzo M. Radiographics\u2013scintigraphic features of autoimmune thyroiditis. Radiographics. 2001;21:957\u201364.","journal-title":"Radiographics"},{"key":"710_CR25","doi-asserted-by":"publisher","first-page":"291","DOI":"10.1016\/S0889-8529(05)70188-2","volume":"30","author":"DA Meier","year":"2001","unstructured":"Meier DA, Kaplan MM. Radioiodine uptake and thyroid scintiscanning. Endocrinol Metab Clin N Am. 2001;30:291\u2013313.","journal-title":"Endocrinol Metab Clin N Am"}],"container-title":["BMC Medical Imaging"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12880-021-00710-4.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s12880-021-00710-4\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12880-021-00710-4.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,1,16]],"date-time":"2023-01-16T08:02:05Z","timestamp":1673856125000},"score":1,"resource":{"primary":{"URL":"https:\/\/bmcmedimaging.biomedcentral.com\/articles\/10.1186\/s12880-021-00710-4"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,11,25]]},"references-count":25,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2021,12]]}},"alternative-id":["710"],"URL":"https:\/\/doi.org\/10.1186\/s12880-021-00710-4","relation":{},"ISSN":["1471-2342"],"issn-type":[{"value":"1471-2342","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,11,25]]},"assertion":[{"value":"11 May 2021","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"19 November 2021","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"25 November 2021","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"This study has been performed in accordance with the Declaration of Helsinki and has been approved by the Institutional Ethics Committee of West China Hospital in Sichuan University and Panzhihua Central Hospital ethics committee. As this study was of retrospective nature, a consent form was waived by the local ethics committee.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics approval and consent to participate"}},{"value":"Not applicable.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}},{"value":"The authors declare there they have no conflict of interest.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"179"}}