Logo-japid
J Adv Periodontol Implant Dent. 2026;18(4): 228-237.
doi: 10.34172/japid.4275
  Abstract View: 11
  PDF Download: 5

Original Article

A comparison of diagnostic accuracy between CNN and transformer AI models for periodontal bone loss detection: A systematic review and meta-analysis

Shaula Nada Aulia 1 ORCID logo, Fabillah Haikal Azizi 1 ORCID logo, Muhammad Hafizh Ash-Shiddiq 1 ORCID logo, Amy Nindia Carabelly 2* ORCID logo, Selviana Rizky Pramitha 3 ORCID logo

1 Undergraduate Program in Dentistry, Faculty of Dentistry, Lambung Mangkurat University, Banjarmasin, Indonesia
2 Department of Oral and Maxillofacial Pathology, Faculty of Dentistry, Lambung Mangkurat University, Banjarmasin, Indonesia
3 Department of Oral Medicine, Faculty of Dentistry, Lambung Mangkurat University, Banjarmasin, Indonesia
*Corresponding Author: Amy Nindia Carabelly, Email: amy.carabelly@ulm.ac.id

Abstract

Introduction: Convolutional neural networks (CNNs) and transformers are artificial intelligence (AI) models used for accurate periodontal bone loss (PBL) diagnosis. This study compared the diagnostic accuracy between CNN and transformer models in detecting PBL.

Methods: We searched ten databases: PubMed, Scopus, Cochrane, Embase, WorldCat, Lens, Google Scholar, Taylor & Francis, ScienceDirect, and ClinicalTrials.gov from 2015 to 2025 to identify studies that used CNN and/or transformer models on human dental radiographs, evaluating diagnostic performance based on area under the curve, accuracy, sensitivity, specificity, correlation matrix, F1-score, precision, and recall. Risk of bias was assessed using PROBAST+AI. Meta-analyses using random-effects models estimated pooled diagnostic odds ratio (DOR), sensitivity, and specificity.

Results: Seventeen studies met the eligibility criteria: 14 evaluated CNN-based models, and 3 evaluated transformer-based models, comprising a total of 26,502 cases and 55,076 controls. The overall risk of bias was low. CNN models demonstrated pooled sensitivity and specificity of 0.85 (95% CI: 0.77–0.90) and 0.88 (95% CI: 0.80–0.93), with a DOR of 41.69 (95% CI: 19.41–89.52). Transformer models showed a sensitivity of 0.87 (95% CI: 0.74–0.94), specificity of 0.80 (95% CI: 0.34–0.97), and DOR of 25.64 (95% CI: 3.47–189.63). Heterogeneity was substantial (I²>98%), with radiograph type and study design identified as significant contributors. CNN results remained robust across sensitivity analyses, while transformer estimates were more sensitive to individual studies.

Conclusion: Current evidence demonstrates greater diagnostic accuracy of CNN-based models for periodontal bone loss detection, while comparative conclusions for transformer-based models remain premature.


First Name
Last Name
Email Address
Comments
Security code


Abstract View: 12

Your browser does not support the canvas element.

PDF Download: 5

Your browser does not support the canvas element.


Submitted: 24 Feb 2026
Revision: 04 Jul 2026
Accepted: 09 Jul 2026
ePublished: 12 Sep 2026
EndNote EndNote

(Enw Format - Win & Mac)

BibTeX BibTeX

(Bib Format - Win & Mac)

Bookends Bookends

(Ris Format - Mac only)

EasyBib EasyBib

(Ris Format - Win & Mac)

Medlars Medlars

(Txt Format - Win & Mac)

Mendeley Web Mendeley Web
Mendeley Mendeley

(Ris Format - Win & Mac)

Papers Papers

(Ris Format - Win & Mac)

ProCite ProCite

(Ris Format - Win & Mac)

Reference Manager Reference Manager

(Ris Format - Win only)

Refworks Refworks

(Refworks Format - Win & Mac)

Zotero Zotero

(Ris Format - Firefox Plugin)