{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,23]],"date-time":"2026-10-23T17:41:28Z","timestamp":1792777288817,"version":"4.5.0"},"reference-count":31,"publisher":"Public Library of Science (PLoS)","issue":"10","license":[{"start":{"date-parts":[[2026,10,23]],"date-time":"2026-10-23T00:00:00Z","timestamp":1792713600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Basic Research Project of Changzhou Medical Center of Nanjing Medical University","award":["CMCB202317"],"award-info":[{"award-number":["CMCB202317"]}],"id":[{"id":"https:\/\/ror.org\/059gcgy73","id-type":"ROR","asserted-by":"crossref"}]}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:sec id=\"sec001\">\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>This study aimed to develop and comparatively evaluate four YOLO nano-scale one-stage object detectors\u2014YOLOv8n, YOLOv10n, YOLO11n, and YOLO12n\u2014for the automated detection of retained root fragments in dental radiographic images, under identical preprocessing, training, and evaluation protocols.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec002\">\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>A curated dataset was organized in the YOLO format and evaluated using a nested 5-fold cross-validation strategy to prevent data leakage. An ablation study was additionally conducted to validate the combined bilateral filtering and histogram equalization (BF\u2009+\u2009HE) preprocessing pipeline. Four YOLO models were fine-tuned from official pretrained weights over 200 epochs. Detection performance was assessed at both the epoch level and the decision-threshold level, reporting the unbiased mean and standard deviation across the independent test folds for precision, sensitivity, F1 score, mean average precision (mAP), and false positives per image (FP\/image).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec003\">\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>In the primary multi-model benchmark, YOLO12n achieved the highest overall detection performance across the held-out test folds, yielding a mean best-epoch F1 score of 0.885\u2009\u00b1\u20090.007, mAP@0.5 of 0.919\u2009\u00b1\u20090.021, and mAP@0.5:0.95 of 0.447\u2009\u00b1\u20090.025 (reaching an optimal-threshold F1 score of 0.888\u2009\u00b1\u20090.016 with 0.346\u2009\u00b1\u20090.052 FP\/image). YOLO11n followed closely as the second-best performing architecture (F1\u2009=\u20090.874\u2009\u00b1\u20090.012, mAP@0.5:0.95\u2009=\u20090.445\u2009\u00b1\u20090.019). Furthermore, the preprocessing ablation study confirmed that the hybrid BF\u2009+\u2009HE pipeline improved detection performance over raw radiographs, elevating YOLO12n's F1 score from 0.856\u2009\u00b1\u20090.013 to 0.888\u2009\u00b1\u20090.016 and mAP@0.5:0.95 from 0.438\u2009\u00b1\u20090.011 to 0.461\u2009\u00b1\u20090.008. All models demonstrated stable convergence behavior.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec004\">\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>Modern YOLO architectures, particularly YOLO12n and YOLO11n, demonstrate technical efficacy for automated RRF localization. The observed performance advantages of YOLO12n are associated with its overall architectural refinements, though dedicated modular ablation studies are warranted to isolate the exact contribution of individual attention mechanisms. These findings demonstrate the technical feasibility of lightweight one-stage object detectors for identifying retained root fragments in dental panoramic radiographs, serving as a baseline for future prospective clinical evaluations.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1371\/journal.pone.0358937","type":"journal-article","created":{"date-parts":[[2026,10,23]],"date-time":"2026-10-23T17:30:59Z","timestamp":1792776659000},"page":"e0358937","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Comparative evaluation of YOLO models for retained root fragment detection"],"prefix":"10.1371","volume":"21","author":[{"given":"Chunyan","family":"Ma","sequence":"first","affiliation":[{"name":"The Changzhou No. 2 People\u2019s Hospital, The Third Affiliated Hospital of Nanjing Medical University, Changzhou Medical Center of Nanjing Medical University","place":["Changzhou, Jiangsu, China"],"department":["Department of Stomatology"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lei","family":"He","sequence":"additional","affiliation":[{"name":"The Changzhou No. 2 People\u2019s Hospital, The Third Affiliated Hospital of Nanjing Medical University, Changzhou Medical Center of Nanjing Medical University","place":["Changzhou, Jiangsu, China"],"department":["Department of Stomatology"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kang","family":"Xia","sequence":"additional","affiliation":[{"name":"The Changzhou No. 2 People\u2019s Hospital, The Third Affiliated Hospital of Nanjing Medical University, Changzhou Medical Center of Nanjing Medical University","place":["Changzhou, Jiangsu, China"],"department":["Department of Stomatology"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaoyuan","family":"Zheng","sequence":"additional","affiliation":[{"name":"The Changzhou No. 2 People\u2019s Hospital, The Third Affiliated Hospital of Nanjing Medical University, Changzhou Medical Center of Nanjing Medical University","place":["Changzhou, Jiangsu, China"],"department":["Department of Information"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-5220-1930","authenticated-orcid":true,"given":"Haibing","family":"Yang","sequence":"additional","affiliation":[{"name":"The Changzhou No. 2 People\u2019s Hospital, The Third Affiliated Hospital of Nanjing Medical University, Changzhou Medical Center of Nanjing Medical University","place":["Changzhou, Jiangsu, China"],"department":["Department of Stomatology"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,10,23]]},"reference":[{"key":"pone.0358937.ref001","doi-asserted-by":"crossref","first-page":"365","DOI":"10.1563\/AAID-JOI-D-14-00023","article-title":"Classification of the alveolar ridge width: Implant-driven treatment considerations for the horizontally deficient alveolar ridges","author":"L Tolstunov","year":"2014","journal-title":"J Oral Implantol"},{"issue":"5","key":"pone.0358937.ref002","doi-asserted-by":"crossref","first-page":"285","DOI":"10.1038\/sj.bdj.2015.147","article-title":"Fractured root tips during dental extractions and retained root fragments. A clinical dilemma?","volume":"218","author":"J Nayyar","year":"2015","journal-title":"Br Dent J"},{"issue":"6","key":"pone.0358937.ref003","doi-asserted-by":"crossref","first-page":"733","DOI":"10.3390\/diagnostics15060733","article-title":"A second residual tooth occurring from one tooth","volume":"15","author":"T Akitomo","year":"2025","journal-title":"Diagnostics"},{"issue":"52","key":"pone.0358937.ref004","article-title":"Management strategies for unanticipated remnants following tooth extraction: A case report","volume":"104","author":"M Liu","year":"2025","journal-title":"Medicine (Baltimore)"},{"issue":"1","key":"pone.0358937.ref005","doi-asserted-by":"crossref","first-page":"20210197","DOI":"10.1259\/dmfr.20210197","article-title":"Current applications and development of artificial intelligence for digital dental radiography","volume":"51","author":"RH Putra","year":"2022","journal-title":"Dentomaxillofac Radiol"},{"issue":"7","key":"pone.0358937.ref006","doi-asserted-by":"crossref","first-page":"769","DOI":"10.1177\/0022034520915714","article-title":"Artificial intelligence in dentistry: Chances and challenges","volume":"99","author":"F Schwendicke","year":"2020","journal-title":"J Dent Res"},{"issue":"5","key":"pone.0358937.ref007","doi-asserted-by":"crossref","first-page":"867","DOI":"10.1016\/j.prosdent.2021.02.010","article-title":"Artificial intelligence applications in restorative dentistry: A systematic review","volume":"128","author":"M Revilla-Le\u00f3n","year":"2022","journal-title":"J Prosthet Dent"},{"key":"pone.0358937.ref008","doi-asserted-by":"crossref","first-page":"103425","DOI":"10.1016\/j.jdent.2020.103425","article-title":"Detecting caries lesions of different radiographic extension on bitewings using deep learning","volume":"100","author":"AG Cantu","year":"2020","journal-title":"J Dent"},{"issue":"5","key":"pone.0358937.ref009","doi-asserted-by":"crossref","first-page":"680","DOI":"10.1111\/iej.13265","article-title":"Evaluation of artificial intelligence for detecting periapical pathosis on cone-beam computed tomography scans","volume":"53","author":"K Orhan","year":"2020","journal-title":"Int Endod J"},{"issue":"1","key":"pone.0358937.ref010","doi-asserted-by":"crossref","first-page":"3840","DOI":"10.1038\/s41598-019-40414-y","article-title":"A deep learning approach to automatic teeth detection and numbering based on object detection in dental periapical films","volume":"9","author":"H Chen","year":"2019","journal-title":"Sci Rep"},{"issue":"1","key":"pone.0358937.ref011","doi-asserted-by":"crossref","first-page":"33304","DOI":"10.1038\/s41598-025-15383-0","article-title":"Comparative analysis of YOLO variants for dental implant brand identification","volume":"15","author":"H Tekis","year":"2025","journal-title":"Sci Rep"},{"issue":"1","key":"pone.0358937.ref012","doi-asserted-by":"crossref","first-page":"7531","DOI":"10.1038\/s41598-020-64509-z","article-title":"Deep learning hybrid method to automatically diagnose periodontal bone loss and stage periodontitis","volume":"10","author":"H-J Chang","year":"2020","journal-title":"Sci Rep"},{"key":"pone.0358937.ref013","article-title":"You only look once: Unified, real-time object detection","author":"J Redmon","year":"2016","journal-title":"IEEE"},{"key":"pone.0358937.ref014","unstructured":"Jocher GCA, Qiu J. YOLO by Ultralytics, GitHub repository. https:\/\/github.com\/ultralytics\/yolo. 2023."},{"key":"pone.0358937.ref015","doi-asserted-by":"crossref","unstructured":"Wang A, Chen H, Liu L, Chen K, Lin Z, Han J, et al. YOLOv10: Real-time end-to-end object detection. 2024. doi: 10.5555\/3737916.3741345","DOI":"10.52202\/079017-3429"},{"key":"pone.0358937.ref016","doi-asserted-by":"crossref","first-page":"105411","DOI":"10.1016\/j.dsp.2025.105411","article-title":"AED-YOLO11: A small object detection model based on YOLO11","volume":"166","author":"X Gong","year":"2025","journal-title":"Digital Signal Processing"},{"issue":"7","key":"pone.0358937.ref017","doi-asserted-by":"crossref","first-page":"2270","DOI":"10.3390\/s25072270","article-title":"YOLO object detection for real-time fabric defect inspection in the textile industry: A review of YOLOv1 to YOLOv11","volume":"25","author":"M Mao","year":"2025","journal-title":"Sensors (Basel)"},{"issue":"1","key":"pone.0358937.ref018","doi-asserted-by":"crossref","first-page":"32546","DOI":"10.1038\/s41598-025-18997-6","article-title":"Advanced real-time detection of acute ischemic stroke using YOLOv12, YOLOv11, and YOLO-NAS: a comparative study for multi-class classification","volume":"15","author":"M El-Geneedy","year":"2025","journal-title":"Sci Rep"},{"key":"pone.0358937.ref019","unstructured":"Tian Y. YOLOv12: Attention-Centric Real-Time Object Detectors. 2023. doi: 10.48550\/arXiv.2502.12524"},{"issue":"17","key":"pone.0358937.ref020","doi-asserted-by":"crossref","first-page":"5389","DOI":"10.3390\/s25175389","article-title":"MCH-YOLOv12: Research on surface defect detection algorithm for aluminum profiles based on improved YOLOv12","volume":"25","author":"Y Sun","year":"2025","journal-title":"Sensors (Basel)"},{"issue":"3","key":"pone.0358937.ref021","doi-asserted-by":"crossref","DOI":"10.1007\/s00521-025-11735-z","article-title":"SSD-YOLOv12: An improved YOLOv12n model for ship detection using SAR data","volume":"38","author":"PT Nguyen","year":"2026","journal-title":"Neural Comput & Applic"},{"issue":"1","key":"pone.0358937.ref022","doi-asserted-by":"crossref","first-page":"508","DOI":"10.1016\/j.jds.2020.06.019","article-title":"Developments, application, and performance of artificial intelligence in dentistry - A systematic review","volume":"16","author":"SB Khanagar","year":"2021","journal-title":"J Dent Sci"},{"issue":"1","key":"pone.0358937.ref023","doi-asserted-by":"crossref","first-page":"358","DOI":"10.1186\/s12903-023-03027-6","article-title":"Artificial intelligence in the diagnosis of dental diseases on panoramic radiographs: A preliminary study","volume":"23","author":"J Zhu","year":"2023","journal-title":"BMC Oral Health"},{"issue":"3","key":"pone.0358937.ref024","doi-asserted-by":"crossref","first-page":"363","DOI":"10.1007\/s11282-021-00572-0","article-title":"Diagnostic charting of panoramic radiography using deep-learning artificial intelligence system","volume":"38","author":"M Ba\u015faran","year":"2022","journal-title":"Oral Radiol"},{"issue":"1","key":"pone.0358937.ref025","doi-asserted-by":"crossref","first-page":"1750","DOI":"10.1186\/s12903-025-07138-0","article-title":"Multi-Regional deep learning models for identifying dental restorations and prosthesis in panoramic radiographs","volume":"25","author":"Z Khurshid","year":"2025","journal-title":"BMC Oral Health"},{"issue":"1","key":"pone.0358937.ref026","doi-asserted-by":"crossref","first-page":"36487","DOI":"10.1038\/s41598-025-21761-5","article-title":"Automated assessment of periapical health based on the radiographic periapical index using YOLOv8, YOLOv11, and YOLOv12 one-stage object detection algorithms","volume":"15","author":"S Saber","year":"2025","journal-title":"Sci Rep"},{"key":"pone.0358937.ref027","doi-asserted-by":"crossref","unstructured":"Dao T, Fu D, Ermon S, Rudra A, R\u00e9 C. FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness. In: Advances in Neural Information Processing Systems 35, 2022. 16344\u201359. https:\/\/doi.org\/10.52202\/068431-1189","DOI":"10.52202\/068431-1189"},{"key":"pone.0358937.ref028","unstructured":"Alif MAR, Hussain M. YOLOv12: A breakdown of the key architectural features. 2025. doi: 10.48550\/arXiv.2502.14740"},{"key":"pone.0358937.ref029","unstructured":"Khanam R, Hussain M. YOLOv11: An overview of the key architectural enhancements. 2024. doi: 10.48550\/arXiv.2410.17725"},{"key":"pone.0358937.ref030","doi-asserted-by":"crossref","unstructured":"Tomasi C, Manduchi R. Bilateral filtering for gray and color images. In: Sixth International Conference on Computer Vision (IEEE Cat. No.98CH36271). 839\u201346. https:\/\/doi.org\/10.1109\/iccv.1998.710815","DOI":"10.1109\/ICCV.1998.710815"},{"issue":"1","key":"pone.0358937.ref031","doi-asserted-by":"crossref","first-page":"247","DOI":"10.1111\/j.1541-0420.2008.01049.x","article-title":"Area under the free-response ROC curve (FROC) and a related summary index","volume":"65","author":"AI Bandos","year":"2009","journal-title":"Biometrics"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0358937","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,23]],"date-time":"2026-10-23T17:31:05Z","timestamp":1792776665000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0358937"}},"subtitle":[],"editor":[{"given":"Xuebo","family":"Zhang","sequence":"first","affiliation":[{"name":"Northwest Normal University","place":["CHINA"]}],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,10,23]]},"references-count":31,"journal-issue":{"issue":"10","published-online":{"date-parts":[[2026,10,23]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0358937","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10,23]]}}}