{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,14]],"date-time":"2026-10-14T07:43:30Z","timestamp":1791963810458,"version":"4.3.4"},"reference-count":35,"publisher":"Public Library of Science (PLoS)","issue":"11","license":[{"start":{"date-parts":[[2025,11,12]],"date-time":"2025-11-12T00:00:00Z","timestamp":1762905600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"\n Purpose<\/jats:title>\n Transvaginal ultrasound (TVUS) is pivotal for diagnosing reproductive pathologies in individuals assigned female at birth, often serving as the primary imaging method for gynecologic evaluation. Despite recent advancements in AI-driven segmentation, its application to gynecological ultrasound still needs further attention. Our study aims to bridge this gap by training and evaluating two state-of-the-art deep learning (DL) segmentation models on TVUS data.<\/jats:p>\n <\/jats:sec>\n \n Materials and methods<\/jats:title>\n An experienced gynecological expert manually segmented the uterus in our TVUS dataset of 124 patients with adenomyosis, comprising still images (n\u2009=\u2009122), video screenshots (n\u2009=\u2009472), and 3D volume screenshots (n\u2009=\u2009452). Two popular DL segmentation models, U-Net and nnU-Net, were trained on the entire dataset, and each imaging type was trained separately. Optimization for U-Net included varying batch size, image resolution, pre-processing, and augmentation. Model performance was measured using the Dice score (DSC).<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n U-Net and nnU-Net had good mean segmentation performances on the TVUS uterus segmentation dataset (0.75 to 0.97 DSC). We observed that training on specific imaging types (still images, video screenshots, 3D volume screenshots) tended to yield better segmentation performance than training on the complete dataset for both models. Furthermore, nnU-Net outperformed the U-Net across all imaging types. Lastly, we report the best results using the U-Net model with limited pre-processing and augmentations.<\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n \n TVUS datasets are well-suited for DL-based segmentation. nnU-Net training was faster and yielded higher segmentation performance; thus, it is recommended over manual U-Net tuning. We also recommend creating TVUS datasets that include only one imaging type and are as clutter-free as possible. The nnU-Net strongly benefited from being trained on 3D volume screenshots in our dataset, likely due to their lack of clutter. Further validation is needed to confirm the robustness of these models on TVUS datasets. Our code is available on\n https:\/\/github.com\/dilaratank\/UtiSeg<\/jats:ext-link>\n .\n <\/jats:p>\n <\/jats:sec>","DOI":"10.1371\/journal.pone.0336237","type":"journal-article","created":{"date-parts":[[2025,11,12]],"date-time":"2025-11-12T18:38:11Z","timestamp":1762972691000},"page":"e0336237","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":3,"title":["Automatic uterus segmentation in transvaginal ultrasound using U-Net and nnU-Net"],"prefix":"10.1371","volume":"20","author":[{"ORCID":"https:\/\/orcid.org\/0009-0006-7252-4753","authenticated-orcid":true,"given":"Dilara","family":"Tank","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bianca G. S.","family":"Schor","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lisa M.","family":"Trommelen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Judith A. F.","family":"Huirne","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Iacer","family":"Calixto","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Robert A.","family":"de Leeuw","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2025,11,12]]},"reference":[{"issue":"5","key":"pone.0336237.ref001","doi-asserted-by":"crossref","first-page":"337","DOI":"10.1590\/0100-3984.2018.0019","article-title":"Transvaginal ultrasound in deep endometriosis: pictorial essay","volume":"52","author":"JGA de Oliveira","year":"2019","journal-title":"Radiol Bras"},{"issue":"4","key":"pone.0336237.ref002","doi-asserted-by":"crossref","first-page":"569","DOI":"10.1016\/j.bpobgyn.2006.01.005","article-title":"Transvaginal ultrasound for diagnosis of adenomyosis: a review","volume":"20","author":"M Dueholm","year":"2006","journal-title":"Best Pract Res Clin Obstet Gynaecol"},{"issue":"1","key":"pone.0336237.ref003","doi-asserted-by":"crossref","first-page":"148","DOI":"10.1016\/j.advms.2022.02.003","article-title":"Diagnostic delay of endometriosis in adults and adolescence-current stage of knowledge","volume":"67","author":"M Wr\u00f3bel","year":"2022","journal-title":"Adv Med Sci"},{"issue":"11","key":"pone.0336237.ref004","doi-asserted-by":"crossref","first-page":"226","DOI":"10.1007\/s10916-018-1088-1","article-title":"Medical Image Analysis using Convolutional Neural Networks: A Review","volume":"42","author":"SM Anwar","year":"2018","journal-title":"J Med Syst"},{"key":"pone.0336237.ref005","first-page":"3711","article-title":"Continual hippocampus segmentation with transformers.","author":"A Ranem","year":"2022","journal-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition"},{"issue":"5","key":"pone.0336237.ref006","article-title":"A Review on Progress in Semantic Image Segmentation and Its Application to Medical Images","volume":"2","author":"MK Kar","year":"2021","journal-title":"SN COMPUT SCI"},{"key":"pone.0336237.ref007","doi-asserted-by":"crossref","first-page":"60","DOI":"10.1016\/j.media.2017.07.005","article-title":"A survey on deep learning in medical image analysis","volume":"42","author":"G Litjens","year":"2017","journal-title":"Med Image Anal"},{"issue":"7","key":"pone.0336237.ref008","doi-asserted-by":"crossref","first-page":"1315","DOI":"10.3390\/cancers16071315","article-title":"Detection and Classification of Hysteroscopic Images Using Deep Learning","volume":"16","author":"D Raimondo","year":"2024","journal-title":"Cancers (Basel)"},{"key":"pone.0336237.ref009","first-page":"1","article-title":"Artificial intelligence applications in endometriosis imaging","author":"S Mittal","year":"2025","journal-title":"Abdom Radiol (NY)"},{"key":"pone.0336237.ref010","first-page":"234","article-title":"U-net: convolutional networks for biomedical image segmentation.","author":"O Ronneberger","year":"2015","journal-title":"International Conference on Medical Image Computing and Computer-Assisted Intervention"},{"issue":"5","key":"pone.0336237.ref011","doi-asserted-by":"crossref","first-page":"1243","DOI":"10.1049\/ipr2.12419","article-title":"Medical image segmentation using deep learning: A survey","volume":"16","author":"R Wang","year":"2022","journal-title":"IET Image Processing"},{"issue":"1","key":"pone.0336237.ref012","doi-asserted-by":"crossref","first-page":"77","DOI":"10.1186\/s41747-023-00388-z","article-title":"Deep learning-based segmentation of multisite disease in ovarian cancer","volume":"7","author":"T Buddenkotte","year":"2023","journal-title":"Eur Radiol Exp"},{"issue":"2","key":"pone.0336237.ref013","doi-asserted-by":"crossref","first-page":"203","DOI":"10.1038\/s41592-020-01008-z","article-title":"nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation","volume":"18","author":"F Isensee","year":"2021","journal-title":"Nat Methods"},{"key":"pone.0336237.ref014","doi-asserted-by":"crossref","first-page":"54310","DOI":"10.1109\/ACCESS.2021.3071301","article-title":"Deep Learning in Medical Ultrasound Image Analysis: A Review","volume":"9","author":"Y Wang","year":"2021","journal-title":"IEEE Access"},{"issue":"10","key":"pone.0336237.ref015","doi-asserted-by":"crossref","first-page":"1832","DOI":"10.1002\/ijc.35092","article-title":"Role of artificial intelligence applied to ultrasound in gynecology oncology: A systematic review","volume":"155","author":"F Moro","year":"2024","journal-title":"Int J Cancer"},{"issue":"3","key":"pone.0336237.ref016","doi-asserted-by":"crossref","first-page":"295","DOI":"10.1002\/uog.29171","article-title":"Application of artificial intelligence to ultrasound imaging for benign gynecological disorders: systematic review","volume":"65","author":"F Moro","year":"2025","journal-title":"Ultrasound Obstet Gynecol"},{"key":"pone.0336237.ref017","doi-asserted-by":"crossref","first-page":"108794","DOI":"10.1016\/j.compbiomed.2024.108794","article-title":"Automatic Segmentation and Alignment of Uterine Shapes from 3D Ultrasound Data","volume":"178","author":"E Bone\u0161","year":"2024","journal-title":"Comput Biol Med"},{"issue":"4","key":"pone.0336237.ref018","doi-asserted-by":"crossref","first-page":"498","DOI":"10.1002\/uog.22122","article-title":"Introduction to artificial intelligence in ultrasound imaging in obstetrics and gynecology","volume":"56","author":"L Drukker","year":"2020","journal-title":"Ultrasound Obstet Gynecol"},{"issue":"3","key":"pone.0336237.ref019","doi-asserted-by":"crossref","first-page":"495","DOI":"10.1002\/jum.16612","article-title":"Grading Sonographic Severity of Adenomyosis: A Pilot Study Assessing Feasibility and Interobserver Reliability","volume":"44","author":"LM Trommelen","year":"2025","journal-title":"J Ultrasound Med"},{"issue":"4","key":"pone.0336237.ref020","doi-asserted-by":"crossref","first-page":"983","DOI":"10.1007\/s10278-022-00620-z","article-title":"The Accuracy and Radiomics Feature Effects of Multiple U-net-Based Automatic Segmentation Models for Transvaginal Ultrasound Images of Cervical Cancer","volume":"35","author":"J Jin","year":"2022","journal-title":"J Digit Imaging"},{"key":"pone.0336237.ref021","first-page":"1","article-title":"Segmentation of ovarian cyst using improved U-Net and hybrid deep learning model","author":"JM Shivaram","year":"2023","journal-title":"Multimed Tools Appl"},{"key":"pone.0336237.ref022","first-page":"993","article-title":"Deep Learning-Based Automatic Endometrium Segmentation and Thickness Measurement for 2D Transvaginal Ultrasound","volume":"2019","author":"S-Y Hu","year":"2019","journal-title":"Annu Int Conf IEEE Eng Med Biol Soc"},{"key":"pone.0336237.ref023","first-page":"1","article-title":"Performance analysis of ultrasound ovarian tumor segmentation using grabcut and FL-SNNM.","author":"K Srilatha","year":"2021","journal-title":"2021 International Conference on Advances in Electrical, Computing, Communication and Sustainable Technologies (ICAECT)"},{"key":"pone.0336237.ref024","first-page":"672","article-title":"Measurement of endometrial thickness using deep neural network with multi-task learning.","author":"J He","year":"2022","journal-title":"Thirteenth International Conference on Graphics and Image Processing (ICGIP 2021)"},{"key":"pone.0336237.ref025","unstructured":"Isensee F, Jaeger PF, Petersen J, Maier-Hein KH. nnU-Net. https:\/\/github.com\/MIC-DKFZ\/nnUNet. Accessed 2024 January 1."},{"key":"pone.0336237.ref026","volume-title":"A method of establishing groups of equal amplitude in plant sociology based on similarity of species content and its application to analyses of the vegetation on Danish commons","author":"TJ S\u00f8rensen","year":"1948"},{"issue":"3","key":"pone.0336237.ref027","doi-asserted-by":"crossref","first-page":"297","DOI":"10.2307\/1932409","article-title":"Measures of the Amount of Ecologic Association Between Species","volume":"26","author":"LR Dice","year":"1945","journal-title":"Ecology"},{"issue":"3","key":"pone.0336237.ref028","doi-asserted-by":"crossref","first-page":"2126","DOI":"10.1109\/TETCI.2024.3377676","article-title":"Advancements in Deep Learning for B-Mode Ultrasound Segmentation: A Comprehensive Review","volume":"8","author":"MY Ansari","year":"2024","journal-title":"IEEE Trans Emerg Top Comput Intell"},{"issue":"1","key":"pone.0336237.ref029","doi-asserted-by":"crossref","first-page":"128","DOI":"10.1186\/s12911-024-02527-x","article-title":"Automatic segmentation of 15 critical anatomical labels and measurements of cardiac axis and cardiothoracic ratio in fetal four chambers using nnU-NetV2","volume":"24","author":"B Liang","year":"2024","journal-title":"BMC Med Inform Decis Mak"},{"issue":"8","key":"pone.0336237.ref030","doi-asserted-by":"crossref","first-page":"4095","DOI":"10.21037\/qims-21-1155","article-title":"Automatic evaluation of endometrial receptivity in three-dimensional transvaginal ultrasound images based on 3D U-Net segmentation","volume":"12","author":"X Wang","year":"2022","journal-title":"Quant Imaging Med Surg"},{"key":"pone.0336237.ref031","doi-asserted-by":"crossref","first-page":"130","DOI":"10.1109\/ECBIOS57802.2023.10218459","article-title":"Application of MP-YOLO for Segmentation and Visualization of Ovarian Ultrasound Imaging.","author":"C-J Wang","year":"2023","journal-title":"2023 IEEE 5th Eurasia Conference on Biomedical Engineering, Healthcare and Sustainability (ECBIOS)"},{"issue":"4","key":"pone.0336237.ref032","doi-asserted-by":"crossref","first-page":"974","DOI":"10.1109\/JBHI.2019.2946092","article-title":"CR-Unet: A Composite Network for Ovary and Follicle Segmentation in Ultrasound Images","volume":"24","author":"H Li","year":"2020","journal-title":"IEEE J Biomed Health Inform"},{"issue":"6","key":"pone.0336237.ref033","article-title":"Early detection and segmentation of ovarian tumor using convolutional neural network with ultrasound imaging","volume":"37","author":"OF Alwan","year":"2023","journal-title":"Rev Intell Artif"},{"key":"pone.0336237.ref034","doi-asserted-by":"crossref","first-page":"1485","DOI":"10.1109\/ULTSYM.2019.8925948","article-title":"Segmentation of gynaecological ultrasound images using different U-Net based approaches.","author":"S Marques","year":"2019","journal-title":"2019 IEEE International Ultrasonics Symposium (IUS)"},{"key":"pone.0336237.ref035","doi-asserted-by":"crossref","first-page":"295","DOI":"10.1007\/978-3-030-27272-2_25","article-title":"Deep learning approaches for gynecological ultrasound image segmentation: A radio-frequency vs B-mode comparison.","author":"C Carvalho","year":"2019","journal-title":"Image Analysis and Recognition: 16th International Conference, ICIAR 2019, Waterloo, ON, Canada, August 27\u201329, 2019, Proceedings, Part II"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0336237","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,12]],"date-time":"2025-11-12T18:38:16Z","timestamp":1762972696000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0336237"}},"subtitle":[],"editor":[{"given":"Paolo","family":"Cazzaniga","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2025,11,12]]},"references-count":35,"journal-issue":{"issue":"11","published-online":{"date-parts":[[2025,11,12]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0336237","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,11,12]]}}}