Digital Periodontal Health Screening and Prevention Program: Implementation of AI-Based Risk Assessment in Diabetic Communities

Triwidawati, Christinne Digital Periodontal Health Screening and Prevention Program: Implementation of AI-Based Risk Assessment in Diabetic Communities. (Unpublished)

[img] Text
Pengabdian Masyarakat Christinne Triwidawati.pdf
Download (208kB)

Abstract

This study aimed to evaluate the accuracy and reliability of digital periodontal assessment using smartphone-based photography combined with machine learning algorithms compared to conventional clinical periodontal examination in community-based screening programs. A total of 320 adults aged 25-65 years from rural communities in Kediri Regency were randomly selected for this cross-sectional study over a 6-month period. Digital periodontal assessment utilized standardized smartphone photography (iPhone 13 Pro) with custom-developed machine learning algorithms trained on 12,000+ periodontal images, while conventional assessment employed standard clinical periodontal examination by certified periodontists. Primary outcomes measured included gingivitis detection accuracy, periodontitis classification accuracy, and inter-rater reliability. Secondary outcomes assessed patient acceptance, cost- effectiveness, and feasibility in community settings. Results showed that digital assessment achieved 94.7% sensitivity and 91.3% specificity for gingivitis detection, and 92.1% sensitivity with 89.8% specificity for periodontitis classification compared to clinical examination. Inter- rater reliability demonstrated excellent agreement (κ=0.89) between digital and clinical assessments. Patient acceptance was high (96.2%), with digital screening requiring 73% less time and 68% lower cost per examination compared to conventional methods. In conclusion, smartphone-based digital periodontal assessment with machine learning demonstrates excellent accuracy and reliability for community-based periodontal screening, offering a cost- effective and accessible alternative to conventional clinical examination. Keywords: Digital Periodontics, Machine Learning, Smartphone Photography, Community Screening, Periodontal Disease, Artificial Intelligence

Item Type: Article
Subjects: R Medicine > RK Dentistry
Divisions: Karya Dosen > Fakultas Kedokteran Gigi
Depositing User: Unnamed user with email rio.mahardiko@gmail.com
Date Deposited: 09 Aug 2025 09:57
Last Modified: 09 Aug 2025 10:04
URI: http://repository.unik-kediri.ac.id/id/eprint/910

Actions (login required)

View Item View Item
Unik Repository is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.