Machine Learning Models for Diagnosing Skeletal Class I and III in German Orthodontic Patients

Machine Learning Models for Diagnosing Skeletal Class I and III in German Orthodontic Patients | 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 Research Article Machine Learning Models for Diagnosing Skeletal Class I and III in German Orthodontic Patients Eva Paddenberg-Schubert, Kareem Midlej, Sebastian Krohn, Agnes Schröder, and 8 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5254525/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 : The precise and efficient diagnosis of an individual’s skeletal class is necessary in orthodontics to ensure correct and stable treatment planning. However, due to several correlations between various anatomic structures, it is difficult to efficiently determine the valid skeletal class. Objectives : The primary outcome of this prospective cross-sectional study was the development of machine learning models for classifying patients as skeletal class I and III. Machine learning regression models were also applied to examine the ability to predict the individualised ANB of Panagiotidis and Witt, using the Wits appraisal parameter. Furthermore, the investigation intended to compare cephalometric variables between skeletal class I and III as well as between age and sex specific subgroups to analyze correlations between cephalometric parameters and to perform Principal Component Analysis (PCA) to identify the most important variables contributing to skeletal class I and III variance. Methods : This study was based on the pre-treatment lateral cephalograms of 509 German orthodontic patients, who were diagnosed as skeletal class I (n = 341) or III (n = 168) according to the individualised ANB of Panagiotidis and Witt. Following descriptive analyses of cephalometric parameters, correlation analyses, and PCA, various machine learning models (RF, CART, KNN, LDA, SVM) and input variables were compared in terms of accuracy, reliability, sensitivity, and specificity in classifying an individual as skeletal class, I or III. Results : Within the same skeletal class, age influenced cephalometric parameters: in skeletal class I, adolescents presented a more horizontal pattern (PFH/AFH, Gonial angle, NL-ML) and prominent mandible (SNB, SN-Pg) than children. In skeletal class III, the degree of sagittal discrepancy between jaw bases was most prominent in adults (ANB: III_Age>21-III _14<Age<20 -1.78°). Comparing skeletal class I and III, the latter had more prognathic mandibles (SNB) and compensated incisors’ inclination (proclination of the upper (+1/NA: 9.01 °), retroinclination of the lower incisors (-1/ML: 8.99°). Among others, a correlation was found between the sagittal (degree of prognathism, SNB) and vertical (inclination, ML-NSL) orientation of the mandible (skeletal class I: p < 0.001, ρ = -0.742; skeletal class III: p < 0.001, ρ = -.665). PCA revealed that the first four principal components explain 93% of the variance in skeletal class I/ III diagnosis and that these parameters had the most influence loading score on the first component- PFH/AFH ratio (0.35), SNB angle (0.35), SN-Pg (0.37), and ML-NSL (-0.35). Evaluating machine learning models, the general model, including all cephalometric parameters, age, and sex, resulted in perfect (1.00) accuracy and kappa scores compared to the gold standard Calculated_ANB with the model's RF and CART. In model 2 the amount of input variables was reduced (Wits, SNB only), but the accuracy (0.88), and kappa (0.73) were still good in the KNN model. Finally, The Wits-appraisal demonstrated a very good (R²=0.61) ability to predict the ANB angle in the machine-learning regression models. the linear regression equation is: . Conclusion: The precise diagnosis of skeletal class I/ III can be simplified by applying the machine learning model KNN with the input variables Wits appraisal and SNB only. This stresses the importance of their correct identification. However, a larger population, considering all skeletal classes, is needed to evaluate the performance of the machine learning model and to improve its performance in terms of kappa and specificity. Finally, Wits appraisal along with gender and age, can predict ANB angle with machine-learning regression models with a perfect fit. Class I class III malocclusion artificial intelligence orthodontic diagnostics individualized treatment planning Full Text Additional Declarations No competing interests reported. Supplementary Files SupplemetaryMaterialsMSIandIIIUKR13102024.docx 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-5254525","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":383953335,"identity":"08ed00df-98bc-42dd-8eef-4558f7115b5f","order_by":0,"name":"Eva Paddenberg-Schubert","email":"","orcid":"","institution":"University Hospital of Regensburg, University of Regensburg","correspondingAuthor":false,"prefix":"","firstName":"Eva","middleName":"","lastName":"Paddenberg-Schubert","suffix":""},{"id":383953336,"identity":"0b0375c1-2cf6-44fb-b978-3133f3134d21","order_by":1,"name":"Kareem Midlej","email":"","orcid":"","institution":"Tel Aviv University","correspondingAuthor":false,"prefix":"","firstName":"Kareem","middleName":"","lastName":"Midlej","suffix":""},{"id":383953337,"identity":"e928e3ce-f396-4e93-a20b-3c52286d76da","order_by":2,"name":"Sebastian Krohn","email":"","orcid":"","institution":"University Hospital of Regensburg, University of Regensburg","correspondingAuthor":false,"prefix":"","firstName":"Sebastian","middleName":"","lastName":"Krohn","suffix":""},{"id":383953338,"identity":"92324540-388c-4e3f-a546-d50994e2da56","order_by":3,"name":"Agnes Schröder","email":"","orcid":"","institution":"University Hospital of Regensburg, University of Regensburg","correspondingAuthor":false,"prefix":"","firstName":"Agnes","middleName":"","lastName":"Schröder","suffix":""},{"id":383953339,"identity":"b1ae607a-dc55-4900-834f-fc0e801a0f52","order_by":4,"name":"Obaida Awadi","email":"","orcid":"","institution":"Center for Dentistry Research and Aesthetics,","correspondingAuthor":false,"prefix":"","firstName":"Obaida","middleName":"","lastName":"Awadi","suffix":""},{"id":383953341,"identity":"a9a82440-8fe7-40ed-9101-fd7b1c48be68","order_by":5,"name":"Samir Masarwa","email":"","orcid":"","institution":"Center for Dentistry Research and Aesthetics,","correspondingAuthor":false,"prefix":"","firstName":"Samir","middleName":"","lastName":"Masarwa","suffix":""},{"id":383953343,"identity":"0677846c-9cc0-40af-a414-4ddd98c6171c","order_by":6,"name":"Iqbal M. Lone","email":"","orcid":"","institution":"Tel Aviv University","correspondingAuthor":false,"prefix":"","firstName":"Iqbal","middleName":"M.","lastName":"Lone","suffix":""},{"id":383953347,"identity":"92631aa0-c46e-4575-bd26-b3979505eb8b","order_by":7,"name":"Osayd Zohud","email":"","orcid":"","institution":"Tel Aviv University","correspondingAuthor":false,"prefix":"","firstName":"Osayd","middleName":"","lastName":"Zohud","suffix":""},{"id":383953348,"identity":"84a26aca-6f45-42fb-901e-9ce43a85a719","order_by":8,"name":"Erika Kuchler","email":"","orcid":"","institution":"University of Bonn","correspondingAuthor":false,"prefix":"","firstName":"Erika","middleName":"","lastName":"Kuchler","suffix":""},{"id":383953349,"identity":"5356cfaa-11c1-4ec9-81d8-1b0ab9ba8ba6","order_by":9,"name":"Nezar Watted","email":"","orcid":"","institution":"Center for Dentistry Research and Aesthetics,","correspondingAuthor":false,"prefix":"","firstName":"Nezar","middleName":"","lastName":"Watted","suffix":""},{"id":383953350,"identity":"b39a1531-b54e-4d5a-8cf7-7df56fae97ad","order_by":10,"name":"Peter Proff","email":"","orcid":"","institution":"University Hospital of Regensburg, University of Regensburg","correspondingAuthor":false,"prefix":"","firstName":"Peter","middleName":"","lastName":"Proff","suffix":""},{"id":383953351,"identity":"631bdc92-8a59-477b-a168-160b0a2e7790","order_by":11,"name":"Fuad A. Iraqi","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA40lEQVRIie3PIevCQBjH8d8QtAysl+ZbuCGIIOJbuTFw5d8sBsMlLWIXxNcwi/mRg62cfSZd/4clMYiooEHBm9Fw33jc5557AJvtB6tJgInh2ykdDMSlO9H8cfVJRBlxxm8ERsLCPMuX5wZPp7xSjBTqE3LMU1i/2Q7W3I91IkCJAtPC/LEeEy12I06chQSqKiAr24VFRxYseC/e5xJ0UWiUk7/bFMmDOKsQNmMFXkrc/0FbJM1wrvuCtrPI9XUgzaQWrXankdedpdo/DI8dz0uVKgoDeYnubwCO/BbYbDab7UNX8tVUPrQ6iCYAAAAASUVORK5CYII=","orcid":"","institution":"University Hospital of Regensburg, University of Regensburg","correspondingAuthor":true,"prefix":"","firstName":"Fuad","middleName":"A.","lastName":"Iraqi","suffix":""}],"badges":[],"createdAt":"2024-10-13 08:53:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5254525/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5254525/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":72652521,"identity":"f8658d9d-5a9e-4281-b5ff-6fc66c13260f","added_by":"auto","created_at":"2024-12-30 19:16:39","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":834565,"visible":true,"origin":"","legend":"","description":"","filename":"MalocclusionIandIIIClean1310.2024.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5254525/v1_covered_755556eb-b2c6-4b5e-8ade-5cbf9f923559.pdf"},{"id":71267173,"identity":"dc82848c-472f-4b37-b243-f7eebffbee8c","added_by":"auto","created_at":"2024-12-12 18:02:23","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":1988348,"visible":true,"origin":"","legend":"","description":"","filename":"SupplemetaryMaterialsMSIandIIIUKR13102024.docx","url":"https://assets-eu.researchsquare.com/files/rs-5254525/v1/ca200aac834e549e0d49f7c5.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Machine Learning Models for Diagnosing Skeletal Class I and III in German Orthodontic Patients","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":"Class I, class III, malocclusion, artificial intelligence, orthodontic diagnostics, individualized treatment planning","lastPublishedDoi":"10.21203/rs.3.rs-5254525/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5254525/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground\u003c/strong\u003e: The precise and efficient diagnosis of an individual’s skeletal class is necessary in orthodontics to ensure correct and stable treatment planning. However, due to several correlations between various anatomic structures, it is difficult to efficiently determine the valid skeletal class.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eObjectives\u003c/strong\u003e: The primary outcome of this prospective cross-sectional study was the development of machine learning models for classifying patients as skeletal class I and III. Machine learning regression models were also applied to examine the ability to predict the individualised ANB of Panagiotidis and Witt, using the Wits appraisal parameter. \u0026nbsp;Furthermore, the investigation intended to compare cephalometric variables between skeletal class I and III as well as between age and sex specific subgroups to analyze correlations between cephalometric parameters and to perform Principal Component Analysis (PCA) to identify the most important variables contributing to skeletal class I and III variance.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods\u003c/strong\u003e: This study was based on the pre-treatment lateral cephalograms of 509 German orthodontic patients, who were diagnosed as skeletal class I (n = 341) or III (n = 168) according to the individualised ANB of Panagiotidis and Witt. Following descriptive analyses of cephalometric parameters, correlation analyses, and PCA, various machine learning models (RF, CART, KNN, LDA, SVM) and input variables were compared in terms of accuracy, reliability, sensitivity, and specificity in classifying an individual as skeletal class, I or III.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e: Within the same skeletal class, age influenced cephalometric parameters: in skeletal class I, adolescents presented a more horizontal pattern (PFH/AFH, Gonial angle, NL-ML) and prominent mandible (SNB, SN-Pg) than children. In skeletal class III, the degree of sagittal discrepancy between jaw bases was most prominent in adults (ANB: III_Age\u0026gt;21-III _14\u0026lt;Age\u0026lt;20 -1.78°). Comparing skeletal class I and III, the latter had more prognathic mandibles (SNB) and compensated incisors’ inclination (proclination of the upper (+1/NA: 9.01 °), retroinclination of the lower incisors (-1/ML: 8.99°). Among others, a correlation was found between the sagittal (degree of prognathism, SNB) and vertical (inclination, ML-NSL) orientation of the mandible (skeletal class I: p \u0026lt; 0.001, ρ = -0.742; skeletal class III: p \u0026lt; 0.001, ρ = -.665). PCA revealed that the first four principal components explain 93% of the variance in skeletal class I/ III diagnosis and that these parameters had the most influence loading score on the first component- PFH/AFH ratio (0.35), SNB angle (0.35), SN-Pg (0.37), and ML-NSL (-0.35). Evaluating machine learning models, the general model, including all cephalometric parameters, age, and sex, resulted in perfect (1.00) accuracy and kappa scores compared to the gold standard Calculated_ANB with the model's RF and CART. In model 2 the amount of input variables was reduced (Wits, SNB only), but the accuracy (0.88), and kappa (0.73) were still good in the KNN model. Finally, The Wits-appraisal demonstrated a very good (R²=0.61) ability to predict the ANB angle in the machine-learning regression models. the linear regression equation is: \u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion:\u003c/strong\u003e The precise diagnosis of skeletal class I/ III can be simplified by applying the machine learning model KNN with the input variables Wits appraisal and SNB only. This stresses the importance of their correct identification. However, a larger population, considering all skeletal classes, is needed to evaluate the performance of the machine learning model and to improve its performance in terms of kappa and specificity. Finally, Wits appraisal along with gender and age, can predict ANB angle with machine-learning regression models with a perfect fit.\u003c/p\u003e","manuscriptTitle":"Machine Learning Models for Diagnosing Skeletal Class I and III in German Orthodontic Patients","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-12-12 18:02:18","doi":"10.21203/rs.3.rs-5254525/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":"d9eb6e56-b143-4ad5-a4f4-5fdd9e6cdde8","owner":[],"postedDate":"December 12th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-12-30T19:08:27+00:00","versionOfRecord":[],"versionCreatedAt":"2024-12-12 18:02:18","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5254525","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5254525","identity":"rs-5254525","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}