Machine Learning and Clustering Analysis of Class II and III Malocclusions
Machine Learning and Clusteranalyse von Klasse-II- und Klasse-III-Malokklusionen
Clinical and Experimental Dental Research · 2026-06 · 12 (3)
DOI: 10.1002/cre2.70384 ↗Auf einen Blick
In 379 deutschen Kieferorthopädie-Patient:innen klassifizierte ein KNN-Modell Skeletal Class II/III mit hoher Genauigkeit. Mit Wits-Appraisal plus SN-Pg angle stieg die Accuracy auf 95,86 %.
Fragestellung
Lassen sich skeletal class II/III mit Machine-Learning-Modellen und Clusteranalyse anhand reduzierter kephalometrischer Parameter korrekt klassifizieren?
Methode
Prospektive multizentrische Beobachtungsstudie mit 379 deutschen Kieferorthopädie-Patient:innen; Vergleich mehrerer ML-Modelle (KNN, RF, LDA, SVM, CART); k-means-Clusteranalyse nach Elbow-Methode.
Ergebnisse
379 Patient:innen: SCIIMO 208 (54,8 %), SCIIIMO 171 (45,1 %). Wits-Appraisal allein: Accuracy 94,72 %; mit SN-Pg angle: 95,86 %. Drei oder vier Cluster waren optimal.
Schlussfolgerung der Autor:innen
Ein reduziertes Parameterset mit Wits-Appraisal und SN-Pg angle ermöglichte eine korrekte Klassifikation. Die explorative Clusteranalyse ergab drei oder vier optimale Cluster und zeigte die Komplexität des Phänotyps.
Kernaussagen
- KNN klassifizierte skeletal class II/III mit hoher Genauigkeit.
- Wits-Appraisal plus SN-Pg angle: Accuracy 95,86 %.
- Clusteranalyse ergab drei oder vier optimale Cluster.
Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.
Kurz gefasst
Multizentrische Studie in der Zeitschrift Clinical and Experimental Dental Research, online erschienen am 01.06.2026. Verfasst von Eva Paddenberg-Schubert und 11 weiteren Autor:innen. Beteiligte Einrichtung: Department of Orthodontics, University Hospital of Regensburg, University of Regensburg, Regensburg, Germany. Hauptthemen laut MEDLINE: Machine Learning, Malocclusion, Angle Class II, Malocclusion, Angle Class III. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 26 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Tel Aviv University, Orthodontic Research Center, University Hospital of Regensburg.
Abstract (Originalsprache)
Objectives
This prospective observational study aims to accurately classify individuals as skeletal class II/III by applying several machine‐learning algorithms. Furthermore, using the k‐means clustering analysis deepened our understanding of the characteristics of skeletal class II/III malocclusion (SCIIMO/SCIIIMO) patients. Materials and Methods This multicenter study consisted of 379 German orthodontic patients. Following their distribution to SCIIMO ( n = 208, 54.8%) or SCIIIMO ( n = 171, 45.1%) by the individualized ANB of Panagiotidis and Witt. Machine‐learning analysis was performed using different models (KNN, RF, LDA, SVM, CART) to find the model with the highest accuracy in correctly classifying skeletal class II/III. After determining the ideal number of clusters using the Elbow approach, k‐means clustering was finally used.
Results
The KNN machine learning model demonstrated the ability of reduced cephalometric parameters to capture much of the same sagittal discriminatory information as the predefined class construct. The results showed that Wits‐appraisal only leads to achieving high accuracy in determining skeletal class (accuracy = 94.72%), while adding SN‐Pg angle improved the accuracy slightly to 95.86%. The clustering analysis showed that three or four clusters were optimal for SCIIMO and SCIIIMO patients and revealed interesting characteristics for each cluster and significance between the differences in all parameters except S‐N and age.
Conclusions
A reduced parameter set that included only the Wits appraisal and SN‐Pg angle enabled a correct classification. In the exploratory clustering analysis, three or four distinct clusters emerged as optimal when analyzing the differences among SCIIMO/SCIIIMO patients, emphasizing the complexity and diversity of malocclusion phenotype.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Orthodontics, University Hospital of Regensburg, University of Regensburg, Regensburg, Germany.
- Department of Clinical Microbiology and Immunology, Faculty of Medicine and Health Sciences, Tel Aviv University, Tel Aviv, Israel.
- Center for Dentistry Research and Aesthetics, Jatt, Israel.
- Department of Oral and Maxillofacial Surgery, Meir Medical Center, Kfar Saba, Israel.
- Faculty of Medicine and Health Sciences, Tel-Aviv University, Tel-Aviv, Israel.
- Department of Orthodontics, University of Bonn, Bonn, Germany.
ORCID-Profile: Fuad A Iraqi
Förderung
- Tel Aviv University
- Orthodontic Research Center
- University Hospital of Regensburg
Zitieren
APA
Paddenberg-Schubert, E., Midlej, K., Krohn, S., Lone, I. M., Zohud, O., Awadi, O., Samir Masarwa, Nashef, A., Kirschneck, C., Watted, N., Proff, P., & Iraqi, F. A. (2026). Machine Learning and Clustering Analysis of Class II and III Malocclusions. Clinical and Experimental Dental Research, 12(3). https://doi.org/10.1002/cre2.70384
Vancouver
Paddenberg-Schubert E, Midlej K, Krohn S, Lone IM, Zohud O, Awadi O, et al. Machine Learning and Clustering Analysis of Class II and III Malocclusions. Clinical and Experimental Dental Research. 2026;12(3). doi:10.1002/cre2.70384
BibTeX
@article{PaddenbergSchubert2026,
author = {Paddenberg-Schubert, Eva and Midlej, Kareem and Krohn, Sebastian and Lone, Iqbal M and Zohud, Osayd and Awadi, Obaida and Samir Masarwa and Nashef, Aysar and Kirschneck, Christian and Watted, Nezar and Proff, Peter and Iraqi, Fuad A},
title = {Machine Learning and Clustering Analysis of Class II and III Malocclusions},
journal = {Clinical and Experimental Dental Research},
year = {2026},
volume = {12},
number = {3},
doi = {10.1002/cre2.70384}
}
EUVIMED