{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,13]],"date-time":"2026-05-13T04:01:10Z","timestamp":1778644870016,"version":"3.51.4"},"reference-count":31,"publisher":"Oxford University Press (OUP)","issue":"1","license":[{"start":{"date-parts":[[2026,1,8]],"date-time":"2026-01-08T00:00:00Z","timestamp":1767830400000},"content-version":"vor","delay-in-days":10,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,12,29]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Intraoperative neuromonitoring\u2014that is, recording of electromyographic signals\u2014is used routinely during (para)thyroid surgery. Surgeons label selected signals to document nerve identity, body side, and time point of stimulation (before or after resection), with a mislabelling rate of 20%. For the purpose of an automated error alert of mislabelled electromyographic signals, the authors developed a multitask one-dimensional convolutional neural network.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>Raw intraoperative neuromonitoring data were corrected using MIONQA software. Labelled electromyographic signals were extracted and metadata (duration of surgery, timing, median electromyographic peak values of actual surgery) were added to each electromyographic wave. Between 150 and 280 extracted features were used to train, validate, and test various convolutional neural networks.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Available raw data from a single centre including 1541 operations with continuous intraoperative nerve monitoring and 508 with intermittent intraoperative nerve monitoring between 2014 and 2024 were used. By repeated adjustments of the model architecture and the number of extracted features, an optimized one-dimensional convolutional neural network was designed. After multiple runs with randomized training (11 414 electromyograms) and test (4891) data, the final optimized convolutional neural network achieved a mean(standard deviation) accuracy of 95.72(0.76)% for correct identification of recurrent laryngeal, vagal, and superior laryngeal nerves; 97.68(0.72)% for correct prediction of the resected body side; and 97.61(0.89)% for correct identification of the stimulation time point (before versus after resection). The receiver operating characteristic curve for classification of the electromyographic peak signals had an excellent area under the curve of 0.993.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>The newly developed convolutional neural network enables accurate automated classification of electromyographic peak signals, facilitating the identification and correction of mislabelled intraoperative nerve monitoring data. Such optimized data quality is essential for artificial intelligence training, enabling neuromonitoring machines to alert the surgeon in the operating theatre of mislabelling. Future studies will aim to include a wider range of clinical scenarios and external data sets, in order to further optimize the existing labelling tool and allow clinical applications.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1093\/bjsopen\/zraf158","type":"journal-article","created":{"date-parts":[[2025,11,12]],"date-time":"2025-11-12T13:05:12Z","timestamp":1762952712000},"source":"Crossref","is-referenced-by-count":1,"title":["Application of deep learning towards automated electromyographic wave classification for neuromonitoring in thyroid and parathyroid surgery"],"prefix":"10.1093","volume":"10","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-4498-7439","authenticated-orcid":false,"given":"Thomas J","family":"Musholt","sequence":"first","affiliation":[{"name":"Section of Endocrine Surgery, Department of General, Visceral and Transplantation Surgery, University Medical Center Mainz , Mainz ,","place":["Germany"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-3398-7283","authenticated-orcid":false,"given":"Petra B","family":"Musholt","sequence":"additional","affiliation":[{"name":"Section of Endocrine Surgery, Department of General, Visceral and Transplantation Surgery, University Medical Center Mainz , Mainz ,","place":["Germany"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0987-8544","authenticated-orcid":false,"given":"Tobias","family":"Kortus","sequence":"additional","affiliation":[{"name":"Chair for Scientific Computing, University of Kaiserslautern-Landau (RPTU) , Kaiserslautern ,","place":["Germany"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"286","published-online":{"date-parts":[[2026,1,8]]},"reference":[{"key":"2026010809303909900_zraf158-B1","doi-asserted-by":"crossref","first-page":"1049","DOI":"10.1007\/s00104-013-2656-z","article-title":"Intraoperative neuromonitoring in thyroid surgery. Recommendations of the Surgical Working Group for Endocrinology","volume":"84","author":"Dralle","year":"2013","journal-title":"Chirurg"},{"key":"2026010809303909900_zraf158-B2","doi-asserted-by":"crossref","first-page":"236","DOI":"10.1001\/jamaoto.2024.4452","article-title":"Progress and outcomes of intraoperative nerve monitoring during thyroidectomy","volume":"151","author":"Hearn","year":"2025","journal-title":"JAMA Otolaryngol Head Neck Surg"},{"key":"2026010809303909900_zraf158-B3","first-page":"S1","article-title":"International neural monitoring study group guideline 2018 part I: staging bilateral thyroid surgery with monitoring loss of signal","volume":"128(Suppl 3)","author":"Schneider","year":"2018","journal-title":"Laryngoscope"},{"key":"2026010809303909900_zraf158-B4","doi-asserted-by":"crossref","first-page":"460","DOI":"10.1002\/hed.26946","article-title":"Clarifying optimal outcome measures in intermittent and continuous laryngeal neuromonitoring","volume":"44","author":"Sinclair","year":"2022","journal-title":"Head Neck"},{"key":"2026010809303909900_zraf158-B5","doi-asserted-by":"crossref","first-page":"1260","DOI":"10.1002\/lary.25807","article-title":"Prospective study of vocal fold function after loss of the neuromonitoring signal in thyroid surgery: the International Neural Monitoring Study Group\u2019s POLT study","volume":"126","author":"Schneider","year":"2016","journal-title":"Laryngoscope"},{"key":"2026010809303909900_zraf158-B6","doi-asserted-by":"crossref","first-page":"499","DOI":"10.1007\/s00423-003-0444-9","article-title":"Validity of intra-operative neuromonitoring signals in thyroid surgery","volume":"389","author":"Thomusch","year":"2004","journal-title":"Langenbecks Arch Surg"},{"key":"2026010809303909900_zraf158-B7","doi-asserted-by":"crossref","first-page":"525","DOI":"10.1002\/lary.27327","article-title":"Prediction of postoperative vocal fold function after intraoperative recovery of loss of signal","volume":"129","author":"Schneider","year":"2019","journal-title":"Laryngoscope"},{"key":"2026010809303909900_zraf158-B8","doi-asserted-by":"crossref","first-page":"675","DOI":"10.1007\/s00423-016-1419-y","article-title":"EMG changes during continuous intraoperative neuromonitoring with sustained recurrent laryngeal nerve traction in a porcine model","volume":"402","author":"Brauckhoff","year":"2017","journal-title":"Langenbecks Arch Surg"},{"key":"2026010809303909900_zraf158-B9","doi-asserted-by":"crossref","first-page":"57","DOI":"10.1016\/j.sopen.2024.12.009","article-title":"A quantitative amplitude of vagus nerve obtained by intraoperative neuromonitoring predicts postoperative vocal cord paralysis among patients in thyroid\/parathyroid surgery as a second option","volume":"23","author":"Katoh","year":"2025","journal-title":"Surg Open Sci"},{"key":"2026010809303909900_zraf158-B10","doi-asserted-by":"crossref","first-page":"zrad051","DOI":"10.1093\/bjsopen\/zrad051","article-title":"Evaluation of intraoperative neuromonitoring (IONM) data with the Mainz IONM quality assurance and analysis tool","volume":"7","author":"Musholt","year":"2023","journal-title":"BJS Open"},{"key":"2026010809303909900_zraf158-B11","first-page":"S1","article-title":"Electrophysiologic recurrent laryngeal nerve monitoring during thyroid and parathyroid surgery: international standards guideline statement","volume":"121(Suppl 1)","author":"Randolph","year":"2011","journal-title":"Laryngoscope"},{"key":"2026010809303909900_zraf158-B12","year":"2024"},{"key":"2026010809303909900_zraf158-B13","author":"Ushey","year":"2024"},{"key":"2026010809303909900_zraf158-B14","doi-asserted-by":"publisher","author":"Abadi","year":"2016","DOI":"10.48550\/arXiv.1605.08695"},{"key":"2026010809303909900_zraf158-B15","author":"Kalinowski","year":"2024"},{"key":"2026010809303909900_zraf158-B16","first-page":"1029","article-title":"Deep learning for echocardiography: introduction for clinicians and future vision. State-of-the-art review","volume":"13","author":"Krittanawong","year":"2023","journal-title":"Life (Basel)"},{"key":"2026010809303909900_zraf158-B17","author":"LeCun","year":"1989"},{"key":"2026010809303909900_zraf158-B18","author":"Ioffe","year":"2015"},{"key":"2026010809303909900_zraf158-B19","first-page":"1929","article-title":"Dropout: a simple way to prevent neural networks from overfitting","volume":"15","author":"Srivastava","year":"2014","journal-title":"J Mach Learn Res"},{"key":"2026010809303909900_zraf158-B20","author":"Gal","year":"2016"},{"key":"2026010809303909900_zraf158-B21","doi-asserted-by":"crossref","first-page":"69","DOI":"10.1515\/cdbme-2021-2018","article-title":"Automated robust interpretation of intraoperative electrophysiological signals\u2014a Bayesian deep learning approach","volume":"7","author":"Kortus","year":"2021","journal-title":"Curr Dir Biomed Eng"},{"key":"2026010809303909900_zraf158-B22","doi-asserted-by":"crossref","first-page":"859","DOI":"10.1080\/01621459.1974.10480219","article-title":"Comparison of several algorithms for computing sample means and variances","volume":"69","author":"Ling","year":"1974","journal-title":"J Am Stat Assoc"},{"key":"2026010809303909900_zraf158-B23","volume-title":"Python 3 Reference Manual","author":"van Rossum","year":"2009"},{"key":"2026010809303909900_zraf158-B24","author":"Kingma"},{"key":"2026010809303909900_zraf158-B25","doi-asserted-by":"crossref","first-page":"77","DOI":"10.1186\/1471-2105-12-77","article-title":"pROC: an open-source package for R and S+ to analyze and compare ROC curves","volume":"12","author":"Robin","year":"2011","journal-title":"BMC Bioinformatics"},{"key":"2026010809303909900_zraf158-B26","doi-asserted-by":"crossref","first-page":"e36427","DOI":"10.2196\/36427","article-title":"Uncertainty estimation in medical image classification: systematic review","volume":"10","author":"Kurz","year":"2022","journal-title":"JMIR Med Inform"},{"key":"2026010809303909900_zraf158-B27","doi-asserted-by":"crossref","first-page":"102830","DOI":"10.1016\/j.artmed.2024.102830","article-title":"Trustworthy clinical AI solutions: a unified review of uncertainty quantification in deep learning models for medical image analysis","volume":"150","author":"Lambert","year":"2024","journal-title":"Artif Intell Med"},{"key":"2026010809303909900_zraf158-B28","doi-asserted-by":"crossref","first-page":"44","DOI":"10.1109\/TMRB.2020.3048255","article-title":"A deep learning model for automated classification of intraoperative continuous EMG","volume":"3","author":"Zha","year":"2021","journal-title":"IEEE Trans Med Robot Bionics"},{"key":"2026010809303909900_zraf158-B29","doi-asserted-by":"crossref","first-page":"11","DOI":"10.1251\/bpo115","article-title":"Techniques of EMG signal analysis: detection, processing, classification and applications","volume":"8","author":"Raez","year":"2006","journal-title":"Biol Proced Online"},{"key":"2026010809303909900_zraf158-B30","doi-asserted-by":"crossref","first-page":"98","DOI":"10.1055\/s-0038-1667083","article-title":"Deep learning on 1-D biosignals: a taxonomy-based survey","volume":"27","author":"Ganapathy","year":"2018","journal-title":"Yearb Med Inform"},{"key":"2026010809303909900_zraf158-B31","doi-asserted-by":"crossref","first-page":"917","DOI":"10.1007\/s10618-019-00619-1","article-title":"Deep learning for time series classification: a review","volume":"33","author":"Fawaz","year":"2019","journal-title":"Data Min Knowl Discov"}],"container-title":["BJS Open"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bjsopen\/article-pdf\/10\/1\/zraf158\/66307389\/zraf158.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bjsopen\/article-pdf\/10\/1\/zraf158\/66307389\/zraf158.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,8]],"date-time":"2026-01-08T14:30:50Z","timestamp":1767882650000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bjsopen\/article\/doi\/10.1093\/bjsopen\/zraf158\/8417287"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12,29]]},"references-count":31,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2025,12,29]]}},"URL":"https:\/\/doi.org\/10.1093\/bjsopen\/zraf158","relation":{},"ISSN":["2474-9842"],"issn-type":[{"value":"2474-9842","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026,2]]},"published":{"date-parts":[[2025,12,29]]},"article-number":"zraf158"}}