Evidence-Based Performance of Artificial Intelligence in Dental Age Assessment: A Systematic Review and Meta-Analysis Using Orthopantomograms

Evidence-Based Performance of Artificial Intelligence in Dental Age Assessment: A Systematic Review and Meta-Analysis Using Orthopantomograms | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Systematic Review Evidence-Based Performance of Artificial Intelligence in Dental Age Assessment: A Systematic Review and Meta-Analysis Using Orthopantomograms Cristiana Palmela Pereira, Ana Rodrigues, Rui Santos, Alexandre Francisco, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8535308/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background: Chronological age estimation based on biological markers plays a crucial role in forensic, legal, and clinical contexts. Among available methods, dental age assessment (DAA) using orthopantomograms (OPGs) is considered one of the most accurate approaches. However, conventional DAA methods rely on expert interpretation, introducing subjectivity and limiting reproducibility. In recent years, artificial intelligence (AI) has emerged as a promising solution to overcome these limitations. This systematic review with meta-analysis aimed to evaluate the performance of AI-based models for dental age assessment using OPGs. Methods: The review protocol followed PRISMA guidelines and was registered in PROSPERO. A total of 27 studies, published between 2020 and 2025, were included, encompassing a wide range of machine learning and deep learning approaches, predominantly convolutional neural networks (CNNs). Results: For chronological age estimation, the pooled root mean squared error (RMSE) was 1.76 years, indicating that AI models can predict age with an average deviation of less than two years from true chronological age. The pooled coefficient of determination (R²) was 0.92, reflecting strong agreement and high explanatory power between predicted and actual ages. In classification tasks related to legal age thresholds, AI models demonstrated high discriminatory performance, with an overall area under the curve (AUC) exceeding 0.90 and a high diagnostic odds ratio (DOR). Hierarchical summary receiver operator characteristic analyses (HSROC) revealed high pooled sensitivity and specificity, with a slightly stronger ability to correctly classify individuals below the legal threshold. Network meta-analyses consistently ranked human expert assessments lower than AI-based models, especially CNN architectures such as EfficientNet, ResNet, and VGG16, although differences were not statistically significant. Conclusions: Despite substantial methodological heterogeneity across studies, the findings suggest that AI-based models for DAA using OPGs have reached a level of accuracy and robustness suitable for forensic and legal applications. AI systems should be regarded as complementary tools, enhancing objectivity, reproducibility, and consistency in age estimation, being good decision-support tools in legal and forensic settings. Dental age assessment Artificial intelligence Orthopantomogram Forensic age estimation Systematic review and meta-analysis Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8535308","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Systematic Review","associatedPublications":[],"authors":[{"id":572965350,"identity":"ca96cd9f-cfb8-43e5-8dda-1db235c82137","order_by":0,"name":"Cristiana Palmela Pereira","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA7klEQVRIiWNgGAWjYNACAwYGfjDJYMHDIEGsFskGsBYJmBYDInQdgGhhIKiFf3bzsQ8fCu7JGx9v3ibB2CYhIz+7+eAHxh1/cGqRuHMseeYMg2LDbWeOlYG08BgARSQYz+C2xUAix5iZxyCBcduNHDOIFokcAyADn5b8z8x/DBLsN89/A9EiPyPH+Ad+LTnMzAwGCYkbJHggWhgg1uHWAvSLMWOPQULyjDNpxRYJ58B+SbNIPGOMUwswxB4z/PiTYNvffnjjjQ9lNvbAEDt84+MOOZxaUOM6AcZIbMCtA0fyYMSnZRSMglEwCkYaAAD6FUyii5MUoAAAAABJRU5ErkJggg==","orcid":"","institution":"Centre of Statistics and its Applications of University of Lisbon (CEAUL)","correspondingAuthor":true,"prefix":"","firstName":"Cristiana","middleName":"Palmela","lastName":"Pereira","suffix":""},{"id":572965351,"identity":"dfe50ff0-c6c7-4498-a2c1-02c9aa8c2b3b","order_by":1,"name":"Ana Rodrigues","email":"","orcid":"","institution":"School of Dental Medicine, University of Zagreb","correspondingAuthor":false,"prefix":"","firstName":"Ana","middleName":"","lastName":"Rodrigues","suffix":""},{"id":572965352,"identity":"5e585af1-9138-481d-ab3b-5ddde64e4edd","order_by":2,"name":"Rui Santos","email":"","orcid":"","institution":"Centre of Statistics and its Applications of University of Lisbon (CEAUL)","correspondingAuthor":false,"prefix":"","firstName":"Rui","middleName":"","lastName":"Santos","suffix":""},{"id":572965353,"identity":"25ede8bd-7f55-421c-99ec-84c8ff643104","order_by":3,"name":"Alexandre Francisco","email":"","orcid":"","institution":"Instituto de Engenharia de Sistemas e Computadores Investigação e Desenvolvimento","correspondingAuthor":false,"prefix":"","firstName":"Alexandre","middleName":"","lastName":"Francisco","suffix":""},{"id":572965355,"identity":"8988c666-7746-449a-a8d0-d40591428a99","order_by":4,"name":"Cátia Vaz","email":"","orcid":"","institution":"Instituto Superior de Engenharia de Lisboa","correspondingAuthor":false,"prefix":"","firstName":"Cátia","middleName":"","lastName":"Vaz","suffix":""},{"id":572965359,"identity":"215dc703-2093-45ab-8531-3ce6e71c4a9e","order_by":5,"name":"Hrvoje Brkić","email":"","orcid":"","institution":"School of Dental Medicine, University of Zagreb","correspondingAuthor":false,"prefix":"","firstName":"Hrvoje","middleName":"","lastName":"Brkić","suffix":""},{"id":572965361,"identity":"86e7ce36-7d78-4076-9cbe-6b54b3554124","order_by":6,"name":"Andreas Schmeling","email":"","orcid":"","institution":"Institute of Legal Medicine of the Munster University Hospital, Polytechnic Institute of Porto","correspondingAuthor":false,"prefix":"","firstName":"Andreas","middleName":"","lastName":"Schmeling","suffix":""},{"id":572965362,"identity":"fb9ee089-3f57-4d53-bb39-5c94a65f9dd7","order_by":7,"name":"João Paulo Martins","email":"","orcid":"","institution":"Centre of Statistics and its Applications of University of Lisbon (CEAUL)","correspondingAuthor":false,"prefix":"","firstName":"João","middleName":"Paulo","lastName":"Martins","suffix":""}],"badges":[],"createdAt":"2026-01-06 23:08:28","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8535308/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8535308/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":100139847,"identity":"1d129a4d-3500-43e1-aa95-eef1bdc80c34","added_by":"auto","created_at":"2026-01-13 11:21:08","extension":"jpg","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":273422,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1PRISMAflowdiagram.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/7ecacc434cc3246c8bff7c91.jpg"},{"id":100139846,"identity":"6360d810-ea88-453d-bc93-24394f4b1b6b","added_by":"auto","created_at":"2026-01-13 11:21:08","extension":"jpg","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":52856,"visible":true,"origin":"","legend":"","description":"","filename":"Figure2AForestplotofrootmeansquareerrorRMSE.Studiesarepresentedfromlowertoptothehigherbottomextractedeffect..jpg","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/f197360c8c74ef685cfff970.jpg"},{"id":100367823,"identity":"17fdf55c-8eea-41d7-9202-28d470c7423c","added_by":"auto","created_at":"2026-01-16 07:57:21","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":138259,"visible":true,"origin":"","legend":"","description":"","filename":"Articlefinal2.docx","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/b53c4fcb145b0a4d1f54f47b.docx"},{"id":100366729,"identity":"4057d044-0cf3-4485-a2a6-df1d884d1812","added_by":"auto","created_at":"2026-01-16 07:56:30","extension":"jpg","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":66178,"visible":true,"origin":"","legend":"","description":"","filename":"Figure2BForestplotofcoefficientofdeterminationR.Studiesarepresentedfromlowertoptothehigherbottomextractedeffect..jpg","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/96ec73716eeaca1630da4e70.jpg"},{"id":100366549,"identity":"f1280a9a-05dd-4076-b7e5-dbdc7457970d","added_by":"auto","created_at":"2026-01-16 07:56:20","extension":"jpg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":43227,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3AForestplotstudiesarebyitsthresholdfromthelowesttoptothehighestthresholdbottomassociatedtotheHSROCmodelofAUCI20.0.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/6e07b1dc3c1033e3b030b7b0.jpg"},{"id":100139851,"identity":"894f032a-cfcf-45ae-bf91-79d7119d5d94","added_by":"auto","created_at":"2026-01-13 11:21:09","extension":"docx","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":17625,"visible":true,"origin":"","legend":"","description":"","filename":"Table1.Searchstringsusedinthestudy.docx","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/4826018b59960bcf2408ff2b.docx"},{"id":100139878,"identity":"6a63860a-e614-4d76-bbe8-c275a47d45ae","added_by":"auto","created_at":"2026-01-13 11:21:12","extension":"jpg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":47764,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3B.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/202d441a24dabe7f81836b23.jpg"},{"id":100366578,"identity":"60baa9ce-61f2-4701-a627-f77067df53ba","added_by":"auto","created_at":"2026-01-16 07:56:22","extension":"docx","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":21176,"visible":true,"origin":"","legend":"","description":"","filename":"Table10.SurfaceunderthecumulativerankingcurveSUCRA.docx","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/f3fcea9e03f56c719bb035b9.docx"},{"id":100367659,"identity":"70da112c-4d85-459f-8433-d879480ab81e","added_by":"auto","created_at":"2026-01-16 07:57:13","extension":"jpg","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":54681,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3C.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/d4da38eb6b45f355d7176a69.jpg"},{"id":100368201,"identity":"64fb816a-5147-4dc7-bd8a-52de5b76f91d","added_by":"auto","created_at":"2026-01-16 07:57:42","extension":"docx","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":84356,"visible":true,"origin":"","legend":"","description":"","filename":"Table11.LeaguetableshowingthecomparisonsforcoefficientofdeterminationR2ofeachmodelclassificationreportedvaluesareFisherztransformed..docx","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/c195244b7ded82c9e0a0ed90.docx"},{"id":100366611,"identity":"d565524c-471e-48ea-b748-1cf000f307e8","added_by":"auto","created_at":"2026-01-16 07:56:23","extension":"jpg","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":52759,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3D.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/0881881efa2b9b9a6c1d94bc.jpg"},{"id":100367848,"identity":"24330e3b-fa34-406b-9499-c21fc788fb91","added_by":"auto","created_at":"2026-01-16 07:57:23","extension":"docx","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":24428,"visible":true,"origin":"","legend":"","description":"","filename":"Table12.SurfaceunderthecumulativerankingcurveSUCRA.docx","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/792ed863ce30ffdeba561379.docx"},{"id":100368132,"identity":"c389af8f-08db-415f-8c62-548b7b954f9a","added_by":"auto","created_at":"2026-01-16 07:57:38","extension":"jpg","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":34367,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3EForestplotstudiesarebyitsthresholdfromthelowesttoptothehighestthresholdbottomassociatedtotheHSROCmodelofDORI298.3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/4371bfeae2ca39602e681ce8.jpg"},{"id":100139848,"identity":"1f9bf63f-49b2-4714-99e9-bea876e9ef17","added_by":"auto","created_at":"2026-01-13 11:21:08","extension":"docx","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":18514,"visible":true,"origin":"","legend":"","description":"","filename":"Table13.LeaveoneoutanalysisforRMSE.docx","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/a781a4cc2b41d9c45ada4a36.docx"},{"id":100139853,"identity":"38c4d078-4b5a-4925-a50c-532f7c6412a6","added_by":"auto","created_at":"2026-01-13 11:21:09","extension":"jpg","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":53302,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3FForestplotstudiesarebyitsthresholdfromthelowesttoptothehighestthresholdbottomassociatedtotheHSROCmodelofaccuracyI298.4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/19d3b8122314b26552394403.jpg"},{"id":100367816,"identity":"7c6087c1-44ac-4055-91bb-63860097cfaf","added_by":"auto","created_at":"2026-01-16 07:57:21","extension":"docx","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":19593,"visible":true,"origin":"","legend":"","description":"","filename":"Table14.LeaveoneoutanalysisforR2.docx","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/74734041dfe0a05f76c8aef4.docx"},{"id":100367831,"identity":"f59217bb-e1d9-4127-bea9-fb5372e8bcce","added_by":"auto","created_at":"2026-01-16 07:57:22","extension":"docx","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":18566,"visible":true,"origin":"","legend":"","description":"","filename":"Table15.LeaveoneoutanalysisforDOR.docx","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/bd1bed0415754fcee679d95e.docx"},{"id":100368293,"identity":"38bcda2d-1bb8-4fd4-bb58-92f1ce22c31d","added_by":"auto","created_at":"2026-01-16 07:57:49","extension":"jpg","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":53030,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3GForestplotstudiesarebyitsthresholdfromthelowesttoptothehighestthresholdbottomassociatedtotheHSROCmodelofF1scoreI298.2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/992488064864008a9cd4f476.jpg"},{"id":100139859,"identity":"7d80b91d-e182-4652-9ae8-bf0284737e63","added_by":"auto","created_at":"2026-01-13 11:21:09","extension":"docx","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":40449,"visible":true,"origin":"","legend":"","description":"","filename":"Table2.Characteristicsoftheincludedstudies.docx","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/1a5b2ddb8f255333bbc5f132.docx"},{"id":100139869,"identity":"e46a5b0b-b466-4991-9a7e-84eb45a17c80","added_by":"auto","created_at":"2026-01-13 11:21:09","extension":"jpg","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":54578,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3HForestplotstudiesarebyitsthresholdfromthelowesttoptothehighestthresholdbottomassociatedtotheHSROCmodelofPrecisionI297.6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/1646cfc9aa6b9c5a6f263011.jpg"},{"id":100139879,"identity":"9b42b84d-4e61-4948-82cf-5e04b6b28f37","added_by":"auto","created_at":"2026-01-13 11:21:13","extension":"docx","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":16508,"visible":true,"origin":"","legend":"","description":"","filename":"Table5.HSROCsummarystatistics..docx","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/231590d9a3f7784a6e0c1f8d.docx"},{"id":100139865,"identity":"6b0e46d1-7a79-4a1e-904a-366774958efa","added_by":"auto","created_at":"2026-01-13 11:21:09","extension":"docx","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":27790,"visible":true,"origin":"","legend":"","description":"","filename":"Table6.SubgroupresultsforDOR.docx","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/b0e6001d1705fc79ea94e5bd.docx"},{"id":100367278,"identity":"16f04ac7-a47d-4a5f-ba4c-33915e20ea4e","added_by":"auto","created_at":"2026-01-16 07:56:55","extension":"jpg","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":29604,"visible":true,"origin":"","legend":"","description":"","filename":"Figure4NetworkdiagramofthediagnosticoddsratiologDORcomparisonsResNet101servesasthecentralhubduetoitsfrequentcomparisonacrossstudies..jpg","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/45e1c06b5854d98ecf87cea5.jpg"},{"id":100366595,"identity":"547b60ac-6418-4a07-a8a3-85fcaacb915d","added_by":"auto","created_at":"2026-01-16 07:56:22","extension":"docx","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":25351,"visible":true,"origin":"","legend":"","description":"","filename":"Table7.LeaguetableshowingthecomparisonsfortheDORofeachmodelclassificationforageclassificationthreshold18years..docx","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/97879f1538b1fafe86c14648.docx"},{"id":100366674,"identity":"e65861f8-fe37-4246-9dc4-9e9c3a9def51","added_by":"auto","created_at":"2026-01-16 07:56:28","extension":"jpg","order_by":24,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":49399,"visible":true,"origin":"","legend":"","description":"","filename":"Figure5NetworkdiagramoftherootmeansquareerrorRMSEcomparisonsResNet101servesasthecentralhubduetoitsfrequentcomparisonacrossstudies..jpg","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/ab82743917edfa50e229f529.jpg"},{"id":100139868,"identity":"9e108b80-bd89-4155-bd53-108a88d61bbe","added_by":"auto","created_at":"2026-01-13 11:21:09","extension":"docx","order_by":25,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":18548,"visible":true,"origin":"","legend":"","description":"","filename":"Table8.Surfaceunderthecumulative.docx","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/5849e7a3c01a06b25408267e.docx"},{"id":100367084,"identity":"bf03d993-b273-4f81-bcfd-9c9ffe2627ca","added_by":"auto","created_at":"2026-01-16 07:56:45","extension":"jpg","order_by":26,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":78467,"visible":true,"origin":"","legend":"","description":"","filename":"Figure6NetworkdiagramofthecoefficientofdeterminationR2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/224df2cfd0cfbc35b380f68a.jpg"},{"id":100367369,"identity":"081172fa-8f8f-48a6-8ea0-28d11cac5abc","added_by":"auto","created_at":"2026-01-16 07:57:02","extension":"docx","order_by":27,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":45940,"visible":true,"origin":"","legend":"","description":"","filename":"Table9.LeaguetableshowingthecomparisonsforrootmeansquareerrorRMSEofeachmodelclassification..docx","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/3f6e61c2d39c877151da0ce6.docx"},{"id":100139877,"identity":"3a43f550-8908-4e33-9cc0-554f7242fb2b","added_by":"auto","created_at":"2026-01-13 11:21:10","extension":"jpg","order_by":28,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":16873,"visible":true,"origin":"","legend":"","description":"","filename":"Figure7AFunnelplotrawandtrimandfilladjustedforRMSEEggersp0.091Beggsp0.761trimandfillnomissingstudies.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/cfebd70bddb75d45bf37822a.jpg"},{"id":100367319,"identity":"34f2f76f-a52b-455e-9151-b465e7f8bfe7","added_by":"auto","created_at":"2026-01-16 07:56:57","extension":"jpg","order_by":29,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":17986,"visible":true,"origin":"","legend":"","description":"","filename":"Figure7BFunnelplotrawandtrimandfilladjustedforREggersp0.364Beggsp0.638trimandfillnomissingstudies.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/afadd87f5be3838e125d6942.jpg"},{"id":100139871,"identity":"b4e9097d-9033-4daf-a3ee-c5bfcb9beca5","added_by":"auto","created_at":"2026-01-13 11:21:09","extension":"jpg","order_by":30,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":17608,"visible":true,"origin":"","legend":"","description":"","filename":"Figure7CFunnelplotrawandtrimandfilladjustedforlogDOREggersp0.514Beggsp0.612trimandfillnomissingstudies.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/8f0884efd06b0734d5f8ec54.jpg"},{"id":100139876,"identity":"d0241349-fd39-4145-8431-93eb8b508353","added_by":"auto","created_at":"2026-01-13 11:21:10","extension":"jpg","order_by":31,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":186982,"visible":true,"origin":"","legend":"","description":"","filename":"Figure8QUADASAIriskofbiasbystudyanddomain.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/1d2d39dccc16f3292f178483.jpg"},{"id":100139875,"identity":"21721dc8-7075-41dd-b619-2191b7dab90a","added_by":"auto","created_at":"2026-01-13 11:21:09","extension":"jpg","order_by":32,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":109038,"visible":true,"origin":"","legend":"","description":"","filename":"Figure9RiskofbiasacrossQUADASAIdomains.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/56f514254fbd34a4122b2d0b.jpg"},{"id":100367077,"identity":"cb05cdc3-6ce9-4059-acf4-c648da2031ed","added_by":"auto","created_at":"2026-01-16 07:56:45","extension":"json","order_by":33,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":9959,"visible":true,"origin":"","legend":"","description":"","filename":"8a84ed825a2347e2accd3a294a459114.json","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1/03922fb6ec117eb12718ed8a.json"},{"id":103398154,"identity":"5bbf4eb6-26a7-4a05-b825-d585a5f08345","added_by":"auto","created_at":"2026-02-25 08:59:00","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1715374,"visible":true,"origin":"","legend":"","description":"","filename":"Articlefinal2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8535308/v1_covered_dd55c4aa-5859-440f-8601-5b34e91efeea.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Evidence-Based Performance of Artificial Intelligence in Dental Age Assessment: A Systematic Review and Meta-Analysis Using Orthopantomograms","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Dental age assessment, Artificial intelligence, Orthopantomogram, Forensic age estimation, Systematic review and meta-analysis","lastPublishedDoi":"10.21203/rs.3.rs-8535308/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8535308/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eBackground: Chronological age estimation based on biological markers plays a crucial role in forensic, legal, and clinical contexts. Among available methods, dental age assessment (DAA) using orthopantomograms (OPGs) is considered one of the most accurate approaches. However, conventional DAA methods rely on expert interpretation, introducing subjectivity and limiting reproducibility. In recent years, artificial intelligence (AI) has emerged as a promising solution to overcome these limitations. This systematic review with meta-analysis aimed to evaluate the performance of AI-based models for dental age assessment using OPGs.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMethods: The review protocol followed PRISMA guidelines and was registered in PROSPERO. A total of 27 studies, published between 2020 and 2025, were included, encompassing a wide range of machine learning and deep learning approaches, predominantly convolutional neural networks (CNNs).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eResults: For chronological age estimation, the pooled root mean squared error (RMSE) was 1.76 years, indicating that AI models can predict age with an average deviation of less than two years from true chronological age. The pooled coefficient of determination (R²) was 0.92, reflecting strong agreement and high explanatory power between predicted and actual ages. In classification tasks related to legal age thresholds, AI models demonstrated high discriminatory performance, with an overall area under the curve (AUC) exceeding 0.90 and a high diagnostic odds ratio (DOR). Hierarchical summary receiver operator characteristic analyses (HSROC) revealed high pooled sensitivity and specificity, with a slightly stronger ability to correctly classify individuals below the legal threshold. Network meta-analyses consistently ranked human expert assessments lower than AI-based models, especially CNN architectures such as EfficientNet, ResNet, and VGG16, although differences were not statistically significant.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eConclusions: Despite substantial methodological heterogeneity across studies, the findings suggest that AI-based models for DAA using OPGs have reached a level of accuracy and robustness suitable for forensic and legal applications. AI systems should be regarded as complementary tools, enhancing objectivity, reproducibility, and consistency in age estimation, being good decision-support tools in legal and forensic settings.\u003c/p\u003e","manuscriptTitle":"Evidence-Based Performance of Artificial Intelligence in Dental Age Assessment: A Systematic Review and Meta-Analysis Using Orthopantomograms","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-01-13 11:20:59","doi":"10.21203/rs.3.rs-8535308/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"5c0c6278-5f36-471c-8302-a243f555434b","owner":[],"postedDate":"January 13th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-02-25T08:56:27+00:00","versionOfRecord":[],"versionCreatedAt":"2026-01-13 11:20:59","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8535308","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8535308","identity":"rs-8535308","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}