{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,8]],"date-time":"2026-10-08T17:55:05Z","timestamp":1791482105041,"version":"4.3.4"},"reference-count":34,"publisher":"Public Library of Science (PLoS)","issue":"10","license":[{"start":{"date-parts":[[2026,10,8]],"date-time":"2026-10-08T00:00:00Z","timestamp":1791417600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Scientific Research Foundation of the Anhui Provincial Education Department","award":["2024AH050970"],"award-info":[{"award-number":["2024AH050970"]}]},{"name":"Scientific Research Foundation of the Anhui Provincial Education Department","award":["KJ2019A0438"],"award-info":[{"award-number":["KJ2019A0438"]}]},{"name":"Scientific Research Foundation of the Anhui Provincial Education Department","award":["SK2020A0255"],"award-info":[{"award-number":["SK2020A0255"]}]}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>Thyroid nodules are a prevalent condition in clinical practice, typically identified via ultrasound imaging with visual assessment by physicians. However, the ambiguous features of malignant nodules, their proximity to surrounding tissues, image noise, and unclear boundaries can lead to variability in diagnostic outcomes among different practitioners. In this study, we initially employed nine established models to segment ultrasound images from two publicly available datasets, TN3K and DDTI. Evaluation metrics including IoU, F1-score, accuracy (Acc), specificity (Spec), and precision (Pre) were utilized to assess the segmentation performance of these models. The experimental results indicate that, given the current data scale, the overall segmentation performance of each model is relatively low. Analysis suggests this is primarily due to the limited number of samples in the medical imaging dataset. During training, the model struggles to fully learn the discriminative features of the lesion area. Particularly during the encoder\u2019s down-sampling stage, the feature expression ability is constrained, affecting the model\u2019s ability to accurately recognize lesion locations. To address this, we enhanced the classic model by incorporating the Residual Network ResNet34, constructing a new thyroid nodule segmentation model. The improved model demonstrated enhanced segmentation performance, with most IoU values on the TN3K dataset exceeding 0.8. However, models trained on the DDTI dataset rarely achieved high performance. Further analysis attributes this performance disparity to differences in the number of training samples. To further enhance model performance, we initialized the improved model with pre-trained ResNet34 weights from the ImageNet dataset, resulting in significant performance improvements. Notably, Res34-PAN performed best across both datasets. On the DDTI dataset, the model achieved an IoU of 0.8778, an F1-score of 0.9349, an accuracy of 0.9872, a specificity of 0.9928, and a precision of 0.9342. To further validate the model design\u2019s rationale, we conducted a systematic ablation experiment. We selected commonly used network structures for encoders and used them as the encoders for nine segmentation models, comparing performance with and without the ImageNet pre-trained weights. The results further confirmed the effectiveness of the proposed method in the thyroid nodule segmentation task.<\/jats:p>","DOI":"10.1371\/journal.pone.0358892","type":"journal-article","created":{"date-parts":[[2026,10,8]],"date-time":"2026-10-08T17:24:30Z","timestamp":1791480270000},"page":"e0358892","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Research on precise segmentation model of thyroid nodules based on deep learning"],"prefix":"10.1371","volume":"21","author":[{"given":"Qingpeng","family":"Zhang","sequence":"first","affiliation":[{"name":"Anhui University of Chinese Medicine","place":["Hefei, Anhui, China"],"department":["School of Medical Information Engineering"]},{"name":"Nanjing University of Aeronautics and Astronautics","place":["Nanjing, Jiangsu, China"],"department":["College of Automation Engineering"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Qinqin","family":"Xiao","sequence":"additional","affiliation":[{"name":"Anhui University of Chinese Medicine","place":["Hefei, Anhui, China"],"department":["School of Medical Information Engineering"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jiacheng","family":"Wu","sequence":"additional","affiliation":[{"name":"Anhui University of Chinese Medicine","place":["Hefei, Anhui, China"],"department":["School of Medical Information Engineering"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4427-0235","authenticated-orcid":true,"given":"Guangming","family":"Shao","sequence":"additional","affiliation":[{"name":"Anhui University of Chinese Medicine","place":["Hefei, Anhui, China"],"department":["School of Medical Information Engineering"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,10,8]]},"reference":[{"issue":"3","key":"pone.0358892.ref001","first-page":"559","article-title":"Thyroid nodules","volume":"67","author":"MJ Welker","year":"2003","journal-title":"Am Fam Phys"},{"issue":"2","key":"pone.0358892.ref002","doi-asserted-by":"crossref","first-page":"329","DOI":"10.1016\/j.mcna.2012.02.002","article-title":"Thyroid nodules","volume":"96","author":"G Popoveniuc","year":"2012","journal-title":"Med Clin North Am"},{"issue":"17","key":"pone.0358892.ref003","doi-asserted-by":"crossref","first-page":"1764","DOI":"10.1056\/NEJMcp031436","article-title":"The thyroid nodule","volume":"351","author":"L Heged\u00fcs","year":"2004","journal-title":"N Engl J Med"},{"issue":"24","key":"pone.0358892.ref004","doi-asserted-by":"crossref","first-page":"2347","DOI":"10.1056\/NEJMcp1415786","article-title":"Thyroid nodules","volume":"373","author":"KD Burman","year":"2015","journal-title":"N Engl J Med"},{"issue":"1","key":"pone.0358892.ref005","doi-asserted-by":"crossref","first-page":"4","DOI":"10.1111\/j.1365-2303.2007.00492.x","article-title":"Thyroid FNAC cytology: can we do it better?","volume":"19","author":"DN Poller","year":"2008","journal-title":"Cytopathology"},{"issue":"1","key":"pone.0358892.ref006","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1186\/s13000-019-0924-z","article-title":"Causes of misdiagnoses by thyroid fine-needle aspiration cytology (FNAC): our experience and a systematic review","volume":"15","author":"Y Zhu","year":"2020","journal-title":"Diagn Pathol"},{"issue":"2","key":"pone.0358892.ref007","doi-asserted-by":"crossref","first-page":"105","DOI":"10.1634\/theoncologist.2007-0212","article-title":"Management of the solitary thyroid nodule","volume":"13","author":"MJ Yeung","year":"2008","journal-title":"Oncologist"},{"issue":"3","key":"pone.0358892.ref008","doi-asserted-by":"crossref","first-page":"360","DOI":"10.1089\/thy.2023.0501","article-title":"Clinical outcomes of thermal ablation re-treatment of benign thyroid nodules: A multicenter study from the italian minimally invasive treatments of the thyroid group","volume":"34","author":"S Bernardi","year":"2024","journal-title":"Thyroid"},{"issue":"1","key":"pone.0358892.ref009","doi-asserted-by":"crossref","DOI":"10.1148\/radiol.240705","article-title":"Evaluation of US Elastography in Thyroid Nodule Diagnosis: The ElaTION Randomized Control Trial","volume":"313","author":"H Mehanna","year":"2024","journal-title":"Radiology"},{"issue":"2","key":"pone.0358892.ref010","doi-asserted-by":"crossref","first-page":"311","DOI":"10.3390\/cancers16020311","article-title":"Integrated diagnostics of thyroid nodules","volume":"16","author":"L Giovanella","year":"2024","journal-title":"Cancers"},{"key":"pone.0358892.ref011","first-page":"3431","article-title":"Fully convolutional networks for semantic segmentation","volume-title":"Proceedings of the IEEE conference on computer vision and pattern recognition","author":"J Long","year":"2015"},{"issue":"4","key":"pone.0358892.ref012","doi-asserted-by":"crossref","first-page":"541","DOI":"10.1162\/neco.1989.1.4.541","article-title":"Backpropagation applied to handwritten zip code recognition","volume":"1","author":"Y LeCun","year":"1989","journal-title":"Neural Comput"},{"key":"pone.0358892.ref013","first-page":"234","article-title":"U-net: Convolutional networks for biomedical image segmentation","volume-title":"Medical image computing and computer-assisted intervention\u2013MICCAI 2015: 18th international conference, Munich, Germany, October 5-9, 2015, proceedings, part III 18","author":"O Ronneberger","year":"2015"},{"key":"pone.0358892.ref014","doi-asserted-by":"crossref","first-page":"3","DOI":"10.1007\/978-3-030-00889-5_1","article-title":"Unet: A nested u-net architecture for medical image segmentation.","volume-title":"Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support: 4th International Workshop, DLMIA 2018, and 8th International Workshop, ML-CDS 2018, Held in Conjunction with MICCAI 2018, Granada, Spain, September 20, 2018, Proceedings 4","author":"Z Zhou","year":"2018"},{"key":"pone.0358892.ref015","first-page":"770","article-title":"Deep residual learning for image recognition.","volume-title":"Proceedings of the IEEE conference on computer vision and pattern recognition","author":"K He","year":"2016"},{"key":"pone.0358892.ref016","doi-asserted-by":"crossref","first-page":"20","DOI":"10.1186\/s12938-015-0014-8","article-title":"Lung segmentation on standard and mobile chest radiographs using oriented Gaussian derivatives filter","volume":"14","author":"WSHM Wan Ahmad","year":"2015","journal-title":"Biomed Eng Online"},{"key":"pone.0358892.ref017","doi-asserted-by":"crossref","first-page":"127445","DOI":"10.1016\/j.neucom.2024.127445","article-title":"Nuclei segmentation using attention aware and adversarial networks","volume":"579","author":"E Goceri","year":"2024","journal-title":"Neurocomputing"},{"key":"pone.0358892.ref018","doi-asserted-by":"crossref","unstructured":"Na S, Guo Y, Jiang F, et al. Segment any cell: A SAM-based auto-prompting fine-tuning framework for nuclei segmentation. 2024.","DOI":"10.1109\/TNNLS.2025.3611322"},{"key":"pone.0358892.ref019","doi-asserted-by":"crossref","first-page":"107858","DOI":"10.1016\/j.bspc.2025.107858","article-title":"FusionLungNet: Multi-scale fusion convolution with refinement network for lung CT image segmentation","volume":"107","author":"S Rezvani","year":"2025","journal-title":"Biomed Signal Process Control"},{"issue":"1","key":"pone.0358892.ref020","article-title":"Lung segmentation from chest X-ray images using Deeplabv3plus-based CNN Model","volume":"13","author":"D Hasan","year":"2024","journal-title":"Indo J Comput Sci"},{"issue":"1","key":"pone.0358892.ref021","doi-asserted-by":"crossref","first-page":"22422","DOI":"10.1038\/s41598-024-72712-5","article-title":"Attention based UNet model for breast cancer segmentation using BUSI dataset","volume":"14","author":"A Sulaiman","year":"2024","journal-title":"Sci Rep"},{"issue":"15","key":"pone.0358892.ref022","doi-asserted-by":"crossref","first-page":"46037","DOI":"10.1007\/s11042-023-17411-5","article-title":"Semantic segmentation of breast cancer images using DenseNet with proposed PSPNet","volume":"83","author":"S Samudrala","year":"2023","journal-title":"Multimed Tools Appl"},{"key":"pone.0358892.ref023","article-title":"Rethinking atrous convolution for semantic image segmentation","author":"LC Chen","year":"2017","journal-title":"arXiv preprint"},{"key":"pone.0358892.ref024","first-page":"833","article-title":"Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation","author":"L-C Chen","year":"2018","journal-title":"Proceedings of the European conference on computer vision (ECCV)"},{"key":"pone.0358892.ref025","first-page":"2117","article-title":"Feature pyramid networks for object detection.","volume-title":"Proceedings of the IEEE conference on computer vision and pattern recognition","author":"TY Lin","year":"2017"},{"key":"pone.0358892.ref026","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/VCIP.2017.8305148","article-title":"Linknet: Exploiting encoder representations for efficient semantic segmentation.","volume-title":"2017 IEEE Visual Communications and Image Processing (VCIP)","author":"A Chaurasia","year":"2017"},{"key":"pone.0358892.ref027","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/TGRS.2022.3179379","article-title":"MANet: Multi-scale aware-relation network for semantic segmentation in aerial scenes","volume":"60","author":"P He","year":"2022","journal-title":"IEEE Trans Geosci Remote Sensing"},{"key":"pone.0358892.ref028","first-page":"2881","article-title":"Pyramid scene parsing network","volume-title":"Proceedings of the IEEE conference on computer vision and pattern recognition","author":"H Zhao","year":"2017"},{"key":"pone.0358892.ref029","first-page":"8759","article-title":"Path aggregation network for instance segmentation","volume-title":"Proceedings of the IEEE conference on computer vision and pattern recognition","author":"S Liu","year":"2018"},{"key":"pone.0358892.ref030"},{"key":"pone.0358892.ref031","doi-asserted-by":"crossref","first-page":"106389","DOI":"10.1016\/j.compbiomed.2022.106389","article-title":"Thyroid region prior guided attention for ultrasound segmentation of thyroid nodules","volume":"155","author":"H Gong","year":"2023","journal-title":"Comput Biol Med"},{"key":"pone.0358892.ref032","doi-asserted-by":"crossref","first-page":"257","DOI":"10.1109\/ISBI48211.2021.9434087","article-title":"Multi-task learning for thyroid nodule segmentation with thyroid region prior","volume-title":"2021 IEEE 18th international symposium on biomedical imaging (ISBI)","author":"H Gong","year":"2021"},{"key":"pone.0358892.ref033"},{"key":"pone.0358892.ref034","first-page":"188","article-title":"An open access thyroid ultrasound image database","volume-title":"10th International symposium on medical information processing and analysis","author":"L Pedraza","year":"2015"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0358892","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,8]],"date-time":"2026-10-08T17:24:40Z","timestamp":1791480280000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0358892"}},"subtitle":[],"editor":[{"given":"Pavan","family":"Kumar","sequence":"first","affiliation":[{"name":"RLBCAU: Rani Lakshmi Bai Central Agricultural University","place":["INDIA"]}],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,10,8]]},"references-count":34,"journal-issue":{"issue":"10","published-online":{"date-parts":[[2026,10,8]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0358892","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10,8]]}}}