EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Assessment of AI-Driven Large Language Models for Orthodontic Aesthetic Scoring Using the IOTN-AC

Ahmet Yıldırım iD, Orhan Cicek iD

Diagnostics · 29.11.2025 · 15 (23) · 3048

DOI: 10.3390/diagnostics15233048 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Diagnostics, online erschienen am 29.11.2025. Verfasst von Ahmet Yıldırım und 1 weiteren Autor:innen. Beteiligte Einrichtung: Department of Orthodontics, Faculty of Dentistry, Zonguldak Bulent Ecevit University, Zonguldak 67600, Türkiye. Die Arbeit wurde bisher 13-mal zitiert (Crossref) und nennt 32 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background/Objectives: The aim of this study was to evaluate the accuracy of aesthetic assessments performed by artificial intelligence (AI)-based large language models (LLMs) using the Aesthetic Component of the Index of Orthodontic Treatment Need (IOTN-AC), which is widely applied to determine the need for orthodontic treatment.

Methods

A total of 150 frontal intraoral photographs from patients in the permanent dentition, scored from 1 to 10 on the IOTN-AC, were assessed by two AI-based LLMs (ChatGPT-5 and ChatGPT-5 Pro). Two experienced clinicians independently scored all photographs, with one evaluator’s scores used as the reference (κ = 0.91, ICC = 0.88). Model performance was analyzed by comparing IOTN-AC scores and treatment need classifications. In addition, performance parameters such as accuracy, precision, specificity, and sensitivity were evaluated. Statistical analyses included Spearman correlation, Cohen’s Kappa, ICC, Mean Absolute Error (MAE), Wilcoxon signed-rank test, and Bland–Altman analysis.

Results

Both models demonstrated positive and significant correlations with the reference values for scoring and classification (p < 0.001). Compared to GPT-5 Pro, the GPT-5 model exhibited superior performance, with a lower error rate (MAE = 1.47) and higher classification accuracy (66.7%). Bland–Altman analysis showed that most predictions fell within the 99% confidence interval, and regression analysis revealed no systematic bias (p > 0.05). Conversely, the models failed to achieve consistently high performance in each of the performance parameters.

Conclusions

The findings revealed that although AI-based LLMs are promising, statistical accuracy alone is insufficient for safe clinical use, and they should demonstrate consistently high performance across all parameters.

Schlagwörter

Stichwörter der Autor:innen

Index of Orthodontic Treatment Needaesthetic componentartificial intelligencelarge language modelocclusal indicesorthodontic treatment need

Autor:innen und Institutionen

  • Department of Orthodontics, Faculty of Dentistry, Zonguldak Bulent Ecevit University, Zonguldak 67600, Türkiye.

ORCID-Profile: Ahmet Yıldırım, Orhan Cicek

Zitieren

APA

Yıldırım, A., & Cicek, O. (2025). Assessment of AI-Driven Large Language Models for Orthodontic Aesthetic Scoring Using the IOTN-AC. Diagnostics, 15(23), 3048. https://doi.org/10.3390/diagnostics15233048

Vancouver

Yıldırım A, Cicek O. Assessment of AI-Driven Large Language Models for Orthodontic Aesthetic Scoring Using the IOTN-AC. Diagnostics. 2025;15(23):3048. doi:10.3390/diagnostics15233048

BibTeX
@article{Yldrm2025,
  author = {Yıldırım, Ahmet and Cicek, Orhan},
  title = {Assessment of AI-Driven Large Language Models for Orthodontic Aesthetic Scoring Using the IOTN-AC},
  journal = {Diagnostics},
  year = {2025},
  volume = {15},
  number = {23},
  pages = {3048},
  doi = {10.3390/diagnostics15233048}
}
Kostenlos starten →