{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,24]],"date-time":"2026-08-24T10:34:53Z","timestamp":1787567693275,"version":"build-2736575974"},"reference-count":47,"publisher":"MDPI AG","issue":"17","license":[{"start":{"date-parts":[[2026,8,23]],"date-time":"2026-08-23T00:00:00Z","timestamp":1787443200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Diagnostics"],"abstract":"<jats:p>Background\/Objectives: Dental age estimation is important in clinical and forensic practice, particularly when skeletal indicators are unavailable or compromised. Machine-learning models often achieve lower regression errors than conventional dental methods; however, whether this translates into better classification performance at legally relevant age thresholds remains unclear. This study compared the Demirjian and Willems methods with several machine learning models for overall accuracy and threshold-specific performance at the jurisdiction-specific ages of 12 and 15 years. Methods: A total of 1384 panoramic radiographs from individuals aged 8.00\u201315.99 years were retrospectively evaluated. The developmental stages of the seven left mandibular permanent teeth and sex were used as model inputs. Linear Regression, Decision Tree, Random Forest, Support Vector Regression, Multilayer Perceptron, Gradient Boosting, and XGBoost were trained using cross-validation and evaluated on an internal holdout set. Performance was assessed using regression errors, age-group-specific bias, sensitivity, specificity, balanced accuracy, and likelihood ratios. Results: Machine-learning models generally produced lower errors than conventional methods. In the holdout set, the lowest mean absolute error was 0.512 years for Support Vector Regression and Gradient Boosting, followed by 0.519 years for XGBoost, compared with 0.649 and 0.654 years for the Willems and Demirjian methods. However, lower regression error did not consistently improve threshold-specific performance. At 12 years, machine learning models increased sensitivity but reduced specificity and positive likelihood ratios relative to Willems. At 15 years, Linear Regression and Random Forest produced no positive predictions, whereas the better-performing models showed results similar to Willems. Conclusions: Lower regression error does not necessarily indicate better forensic threshold-specific classification performance. Dental age-estimation models should therefore be validated using threshold-specific likelihood ratios, classification metrics, and age-group-specific bias in addition to overall prediction errors.<\/jats:p>","DOI":"10.3390\/diagnostics16172690","type":"journal-article","created":{"date-parts":[[2026,8,24]],"date-time":"2026-08-24T09:20:02Z","timestamp":1787563202000},"page":"2690","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Does Lower Regression Error Mean Stronger Forensic Evidence? Machine Learning Regression Versus Demirjian and Willems Methods for Dental Age Estimation at 12- and 15-Year Legal Thresholds"],"prefix":"10.3390","volume":"16","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-3568-2369","authenticated-orcid":false,"given":"Mustafa","family":"Do\u011fan","sequence":"first","affiliation":[{"name":"Department of Forensic Medicine, Faculty of Medicine, Ni\u011fde \u00d6mer Halisdemir University, 51000 Nigde, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6614-9903","authenticated-orcid":false,"given":"Muhammed Emin","family":"Parlak","sequence":"additional","affiliation":[{"name":"Department of Forensic Medicine, Faculty of Medicine, Ni\u011fde \u00d6mer Halisdemir University, 51000 Nigde, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kadir Sezer","family":"Ko\u00e7ak","sequence":"additional","affiliation":[{"name":"Ni\u011fde Forensic Medicine Branch Office, The Council of Forensic Medicine, Ministry of Justice, 51000 Nigde, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7369-6083","authenticated-orcid":false,"given":"Yasin","family":"Etli","sequence":"additional","affiliation":[{"name":"Department of Forensic Medicine, Faculty of Medicine, Van Yuzuncu Yil University, 65000 Van, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bora","family":"\u00d6zdemir","sequence":"additional","affiliation":[{"name":"Department of Forensic Medicine, Faculty of Medicine, Ni\u011fde \u00d6mer Halisdemir University, 51000 Nigde, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Katibe Tu\u011f\u00e7e","family":"Temur","sequence":"additional","affiliation":[{"name":"Department of Oral and Maxillofacial Radiology, Faculty of Dentistry, Ni\u011fde \u00d6mer Halisdemir University, 51000 Nigde, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,8,23]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"1490","DOI":"10.1177\/00220345630420062701","article-title":"Age Variation of Formation Stages for Ten Permanent Teeth","volume":"42","author":"Moorrees","year":"1963","journal-title":"J. Dent. Res."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"91","DOI":"10.5624\/isd.20221226","article-title":"Dental Age Estimation Using Cone-Beam Computed Tomography: A Systematic Review and Meta-Analysis","volume":"53","author":"Yousefi","year":"2023","journal-title":"Imaging Sci. Dent."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"47","DOI":"10.1016\/j.scijus.2020.09.004","article-title":"Minor or Adult? Introducing Decision Analysis in Forensic Age Estimation","volume":"61","author":"Sironi","year":"2021","journal-title":"Sci. Justice"},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"1459","DOI":"10.1007\/s00414-024-03162-x","article-title":"Efficacy of the Methods of Age Determination Using Artificial Intelligence in Panoramic Radiographs\u2014A Systematic Review","volume":"138","author":"Escobar","year":"2024","journal-title":"Int. J. Leg. Med."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"315","DOI":"10.1007\/s00414-010-0515-8","article-title":"Accuracy of Cameriere, Haavikko, and Willems Radiographic Methods on Age Estimation on Bosnian-Herzegovian Children Age Groups 6\u201313","volume":"125","author":"Cameriere","year":"2011","journal-title":"Int. J. Leg. Med."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"178","DOI":"10.1016\/j.forsciint.2006.05.016","article-title":"Age Estimation","volume":"165","author":"Schmeling","year":"2007","journal-title":"Forensic Sci. Int."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"130","DOI":"10.1016\/j.jflm.2017.09.003","article-title":"Is Willems Method Universal for Age Estimation: A Systematic Review and Meta-Analysis","volume":"52","author":"Wang","year":"2017","journal-title":"J. Forensic Leg. Med."},{"key":"ref_8","first-page":"211","article-title":"A New System of Dental Age Assessment","volume":"45","author":"Demirjian","year":"1973","journal-title":"Hum. Biol."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"893","DOI":"10.1520\/JFS15064J","article-title":"Dental Age Estimation in Belgian Children: Demirjian\u2019s Technique Revisited","volume":"46","author":"Willems","year":"2001","journal-title":"J. Forensic Sci."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"753","DOI":"10.1007\/s00167-022-06896-6","article-title":"Machine Learning and Conventional Statistics: Making Sense of the Differences","volume":"30","author":"Ley","year":"2022","journal-title":"Knee Surg. Sports Traumatol. Arthrosc."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"171","DOI":"10.1007\/s10522-017-9683-y","article-title":"A Review of Supervised Machine Learning Applied to Ageing Research","volume":"18","author":"Fabris","year":"2017","journal-title":"Biogerontology"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"10","DOI":"10.1007\/s12024-025-01042-3","article-title":"Dental Age Estimation by Comparing Demirjian\u2019s Method and Machine Learning in Southeast Brazilian Youth","volume":"22","author":"Abuabara","year":"2025","journal-title":"Forensic Sci. Med. Pathol."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"665","DOI":"10.1007\/s00414-020-02489-5","article-title":"Comparison of Different Machine Learning Approaches to Predict Dental Age Using Demirjian\u2019s Staging Approach","volume":"135","author":"Galibourg","year":"2021","journal-title":"Int. J. Leg. Med."},{"key":"ref_14","first-page":"135","article-title":"Dental Age Estimation in Northern Chinese Han Children and Adolescents Using Demirjian\u2019s Method Combined with Machine Learning Algorithms","volume":"40","author":"Guo","year":"2024","journal-title":"Fa Yi Xue Za Zhi"},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Esan, T.A., Yengopal, V., and Schepartz, L.A. (2017). The Demirjian versus the Willems Method for Dental Age Estimation in Different Populations: A Meta-Analysis of Published Studies. PLoS ONE, 12.","DOI":"10.1371\/journal.pone.0186682"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"680","DOI":"10.1111\/1556-4029.13926","article-title":"Status of Mandibular Third Molar Development as Evidence in Legal Age Threshold Cases","volume":"64","author":"Konigsberg","year":"2019","journal-title":"J. Forensic Sci."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"1451","DOI":"10.1007\/s00414-026-03721-4","article-title":"Dental Age Estimation: A Scoping Review Comparing the Manual Application of the Demirjian Method and Artificial Intelligence Modalities","volume":"140","author":"Baylis","year":"2026","journal-title":"Int. J. Leg. Med."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"4974","DOI":"10.1109\/TPAMI.2025.3552368","article-title":"Systematic Bias of Machine Learning Regression Models and Correction","volume":"47","author":"Lee","year":"2025","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref_19","first-page":"44","article-title":"Forensic Age Estimation","volume":"113","author":"Schmeling","year":"2016","journal-title":"Dtsch. Arztebl. Int."},{"key":"ref_20","first-page":"2","article-title":"Recommendations for Quality Assurance: Dental Age Assessment in Living Individuals","volume":"44","author":"Pereira","year":"2026","journal-title":"J. Forensic Odontostomatol."},{"key":"ref_21","unstructured":"European Union Agency for Asylum (EUAA) (2026, August 16). Practical Guide on Age Assessment. Available online: https:\/\/www.euaa.europa.eu\/publications\/practical-guide-age-assessment-0."},{"key":"ref_22","first-page":"9","article-title":"A Review of the Most Commonly Used Dental Age Estimation Techniques","volume":"19","author":"Willems","year":"2001","journal-title":"J. Forensic Odontostomatol."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"101991","DOI":"10.1016\/j.legalmed.2021.101991","article-title":"Accuracy of the London Atlas, Haavikko\u2019s Method and Cameriere\u2019s European Formula of Dental Age Estimation in Turkish Children","volume":"54","author":"Sezer","year":"2022","journal-title":"Leg. Med."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"111654","DOI":"10.1016\/j.forsciint.2023.111654","article-title":"Reliability and Limitations of Permanent Tooth Staging Techniques","volume":"346","author":"Davies","year":"2023","journal-title":"Forensic Sci. Int."},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Muzzall, E. (2021). A Novel Ensemble Machine Learning Approach for Bioarchaeological Sex Prediction. Technologies, 9.","DOI":"10.3390\/technologies9020023"},{"key":"ref_26","unstructured":"Mechelli, A., and Vieira, S. (2019). Support Vector Regression. Machine Learning: Methods and Applications to Brain Disorders, Academic Press."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"103490","DOI":"10.1016\/j.compbiomed.2019.103490","article-title":"A Trial on Artificial Neural Networks in Predicting Sex through Bone Length Measurements on the First and Fifth Phalanges and Metatarsals","volume":"115","author":"Turan","year":"2019","journal-title":"Comput. Biol. Med."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"1189","DOI":"10.1214\/aos\/1013203451","article-title":"Greedy Function Approximation: A Gradient Boosting Machine","volume":"29","author":"Friedman","year":"2001","journal-title":"Ann. Stat."},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Chen, T., and Guestrin, C. (2016). XGBoost: A Scalable Tree Boosting System. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD \u201916), San Francisco, CA, USA, 13\u201317 August 2016, Association for Computing Machinery.","DOI":"10.1145\/2939672.2939785"},{"key":"ref_30","unstructured":"Liashchynskyi, P., and Liashchynskyi, P. (2019). Grid Search, Random Search, Genetic Algorithm: A Big Comparison for NAS. arXiv."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"85","DOI":"10.1016\/j.jas.2016.04.004","article-title":"Machine Learning-Based Approaches for Predicting Stature from Archaeological Skeletal Remains Using Long Bone Lengths","volume":"69","author":"Czibula","year":"2016","journal-title":"J. Archaeol. Sci."},{"key":"ref_32","first-page":"102","article-title":"Use and Misuse of Machine Learning in Anthropology","volume":"2","author":"Calder","year":"2022","journal-title":"IEEE BITS Inf. Theory Mag."},{"key":"ref_33","unstructured":"(2026, July 11). Mevzuat Bilgi Sistemi. 5237 Say\u0131l\u0131 T\u00fcrk Ceza Kanunu, Available online: https:\/\/www.mevzuat.gov.tr\/mevzuat?MevzuatNo=5237&MevzuatTur=1&MevzuatTertip=5."},{"key":"ref_34","unstructured":"European Network of Forensic Science Institutes (ENFSI) (2026, July 11). ENFSI Guideline for Evaluative Reporting in Forensic Science. Available online: https:\/\/enfsi.eu\/documents\/forensic-guidelines\/."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"373","DOI":"10.1016\/j.jflm.2013.03.015","article-title":"The French-Canadian Data Set of Demirjian for Dental Age Estimation: A Systematic Review and Meta-Analysis","volume":"20","author":"Jayaraman","year":"2013","journal-title":"J. Forensic Leg. Med."},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"80","DOI":"10.26650\/eor.20221057985","article-title":"Accuracy of Age Estimation with Demirjian and Nolla Methods in Eastern Turkish Children Aged 3\u201317 Years Old","volume":"56","author":"Duruk","year":"2022","journal-title":"Eur. Oral Res."},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"122","DOI":"10.1016\/j.jflm.2017.08.017","article-title":"Willems Method of Dental Age Estimation in Children: A Systematic Review and Meta-Analysis","volume":"52","author":"Sehrawat","year":"2017","journal-title":"J. Forensic Leg. Med."},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"109912","DOI":"10.1016\/j.forsciint.2019.109912","article-title":"A Comparison of the Accuracy of Willems\u2019 and Cameriere\u2019s Methods Based on Panoramic Radiography","volume":"302","author":"Ozveren","year":"2019","journal-title":"Forensic Sci. Int."},{"key":"ref_39","doi-asserted-by":"crossref","unstructured":"G\u00fcm\u00fc\u015fbo\u011fa, Z.\u015e., Duruk, G., \u00d6zdemir, E., Aydan, T., and K\u0131rg\u0131n, A.\u015e. (2024). An Investigation of the Validity of the Bedek Models and Cameriere-European Formula Used in Dental Age Prediction in Turkish Children. BMC Oral Health, 24.","DOI":"10.1186\/s12903-024-05166-w"},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"3623","DOI":"10.1007\/s10278-025-01769-z","article-title":"Improving Chronological Age Estimation in Children Using the Demirjian Method Enhanced with Transformer and Regression Models","volume":"39","author":"Simsek","year":"2026","journal-title":"J. Imaging Inform. Med."},{"key":"ref_41","doi-asserted-by":"crossref","unstructured":"Lutescu, D.A., Manole, C.C., \u021a\u00e2ncu, A.M.C., Ilinca, R., Chau, R.C.W., Felszeghy, S., and Didilescu, A.C. (2026). Can Artificial Intelligence in Orthopantomography Advance Dental Diagnostics through Automated Image Analysis?. Front. Radiol., 6.","DOI":"10.3389\/fradi.2026.1701356"},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"21","DOI":"10.1002\/cbm.70024","article-title":"The Minimum Age of Criminal Responsibility Internationally\u2014History, Systems and the Future","volume":"36","author":"Delmage","year":"2026","journal-title":"Crim. Behav. Ment. Health"},{"key":"ref_43","unstructured":"United Nations Committee on the Rights of the Child (2019). General Comment No. 24 (2019) on Children\u2019s Rights in the Child Justice System, United Nations. CRC\/C\/GC\/24."},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"102885","DOI":"10.1016\/j.legalmed.2026.102885","article-title":"Artificial Intelligence in Forensic Age Assessment: A Systematic Review","volume":"84","author":"Bugelli","year":"2026","journal-title":"Leg. Med."},{"key":"ref_45","doi-asserted-by":"crossref","first-page":"112353","DOI":"10.1016\/j.forsciint.2024.112353","article-title":"Do Machine Learning Methods Solve the Main Pitfall of Linear Regression in Dental Age Estimation?","volume":"367","author":"Faragalli","year":"2025","journal-title":"Forensic Sci. Int."},{"key":"ref_46","doi-asserted-by":"crossref","first-page":"112979","DOI":"10.1016\/j.forsciint.2026.112979","article-title":"Bias-Reduced Dental Age Estimation in Children: Machine-Learning Evaluation and a Novel Constrained Random Forest Adaptation of the Cameriere Method","volume":"385","author":"Parlak","year":"2026","journal-title":"Forensic Sci. Int."},{"key":"ref_47","doi-asserted-by":"crossref","first-page":"1661","DOI":"10.1007\/s00414-025-03452-y","article-title":"Comparison of Different Dental Age Estimation Methods with Deep Learning: Willems, Cameriere-European, London Atlas","volume":"139","author":"Yavuz","year":"2025","journal-title":"Int. J. Leg. Med."}],"container-title":["Diagnostics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/17\/2690\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,24]],"date-time":"2026-08-24T09:35:09Z","timestamp":1787564109000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/17\/2690"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,23]]},"references-count":47,"journal-issue":{"issue":"17","published-online":{"date-parts":[[2026,9]]}},"alternative-id":["diagnostics16172690"],"URL":"https:\/\/doi.org\/10.3390\/diagnostics16172690","relation":{},"ISSN":["2075-4418"],"issn-type":[{"value":"2075-4418","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,23]]}}}