EUVIMEDEUVIMED entdecken
Open Access

The efficacy of supervised learning and semi-supervised learning in diagnosis of impacted third molar on panoramic radiographs through artificial intelligence model

Ji-Youn Kim, Se Hoon Kahm, Seok Yoo, Soo-Mi Bae, Ji-Eun Kang, Sang Hwa Lee iD

Dentomaxillofacial Radiology · 2023

DOI: 10.1259/dmfr.20230030 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Dentomaxillofacial Radiology, 2023. Verfasst von Ji-Youn Kim und 5 weiteren Autor:innen. Beteiligte Einrichtung: Division of Oral & Maxillofacial Surgery, Department of Dentistry, St. Vincent's Hospital, College of Medicine, The Catholic University of Korea, Seo…. Hauptthemen laut MEDLINE: Molar, Third, Tooth, Impacted. Die Arbeit wurde bisher 15-mal zitiert (Crossref) und nennt 22 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objectives

The aim of the study was to evaluate the efficacy of traditional supervised learning (SL) and semi-supervised learning (SSL) in the classification of mandibular third molars (Mn3s) on panoramic images. The simplicity of preprocessing step and the outcome of the performance of SL and SSL were analyzed.

Methods

Total 1625 Mn3s cropped images from 1000 panoramic images were labeled for classifications of the depth of impaction (D class), spatial relation with adjacent second molar (S class), and relationship with inferior alveolar nerve canal (N class). For the SL model, WideResNet (WRN) was applicated and for the SSL model, LaplaceNet (LN) was utilized.

Results

In the WRN model, 300 labeled images for D and S classes, and 360 labeled images for N class were used for training and validation. In the LN model, only 40 labeled images for D, S, and N classes were used for learning. The F1 score were 0.87, 0.87, and 0.83 in WRN model, 0.84, 0.94, and 0.80 for D class, S class, and N class in the LN model, respectively.

Conclusions

These results confirmed that the LN model applied as SSL, even utilizing a small number of labeled images, demonstrated the satisfactory of the prediction accuracy similar to that of the WRN model as SL.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Molar, ThirdTooth, ImpactedHumansRadiography, PanoramicArtificial IntelligenceMandibleCone-Beam Computed TomographySupervised Machine Learning

Stichwörter der Autor:innen

deep learningpanoramic radiographysemi-supervised learningsupervised machine learningthird molar

Autor:innen und Institutionen

  • Division of Oral & Maxillofacial Surgery, Department of Dentistry, St. Vincent's Hospital, College of Medicine, The Catholic University of Korea, Seoul, Republic of Korea.
  • Department of Dentistry, Eunpyeong St. Mary's Hospital, College of Medicine, The Catholic University of Korea, Seoul, Republic of Korea.
  • AI Business Headquarters, Unidocs Inc., Seoul, South Korea.
  • Department of Artificial Intelligence, Graduate school, Korea University, Seoul, South Korea.
  • JINHAKapply Corp., Seoul, South Korea.

ORCID-Profile: Sang Hwa Lee

Zitieren

APA

Kim, J. Y., Kahm, S. H., Yoo, S., Bae, S. M., Kang, J. E., & Lee, S. H. (2023). The efficacy of supervised learning and semi-supervised learning in diagnosis of impacted third molar on panoramic radiographs through artificial intelligence model. Dentomaxillofacial Radiology. https://doi.org/10.1259/dmfr.20230030

Vancouver

Kim JY, Kahm SH, Yoo S, Bae SM, Kang JE, Lee SH. The efficacy of supervised learning and semi-supervised learning in diagnosis of impacted third molar on panoramic radiographs through artificial intelligence model. Dentomaxillofacial Radiology. 2023. doi:10.1259/dmfr.20230030

BibTeX
@article{Kim2023,
  author = {Kim, Ji-Youn and Kahm, Se Hoon and Yoo, Seok and Bae, Soo-Mi and Kang, Ji-Eun and Lee, Sang Hwa},
  title = {The efficacy of supervised learning and semi-supervised learning in diagnosis of impacted third molar on panoramic radiographs through artificial intelligence model},
  journal = {Dentomaxillofacial Radiology},
  year = {2023},
  doi = {10.1259/dmfr.20230030}
}
Kostenlos starten →