Related Experiment Video
Updated: Sep 2, 2026
Precision of In Vivo Quantitative Tooth Wear Measurement Using Intra-Oral Scans
Published on: July 12, 2022
Clinical validation of a web-based decision support system for periodontal diagnosis, risk assessment, and recall
Aruna Nautiyal1, Tanya Attre1, Sonali Singh2
1ITS CDSR, Ghaziabad, India.
Purpose:
Application of the 2017 American Academy of Periodontology/European Federation of Periodontology (AAP/EFP) classification requires integration of multiple clinical, radiographic, and risk-related parameters, which may be challenging and time-consuming in routine clinical practice. Digital decision-support systems may assist clinicians by improving diagnostic consistency and workflow efficiency. Therefore, the aim of the current study is to evaluate the diagnostic agreement, accuracy, and clinical efficiency of a web-based automated decision-support system for periodontal diagnosis, risk assessment, and recall planning in comparison with gold-standard clinician assessment.
Methods:
This two-phase study included Rule-Based Algorithm Development and Verification using 2000 anonymized periodontal records and prospective clinical validation in 328 consecutive patients. Automated outputs for periodontal staging, grading, periodontal risk assessment, and supportive periodontal therapy recall recommendations were compared with independent clinician diagnosis. Diagnostic agreement was assessed using Cohen's kappa statistics, while sensitivity, specificity, predictive values, and diagnostic time were also analyzed.
Results:
The automated system demonstrated moderate agreement with clinician diagnosis for periodontal staging and grading, with higher levels of agreement observed for periodontal risk categorization and recall interval recommendations. Sensitivity and negative predictive value for identification of advanced periodontitis (Stage III-IV) were high, while most diagnostic discrepancies occurred between adjacent diagnostic categories. Use of the system resulted in a significant reduction in diagnostic time compared with conventional clinical assessment (p < 0.0001).
Conclusions:
The web-based decision-support system showed clinically acceptable agreement with clinician diagnosis and improved diagnostic efficiency. It may be used as an adjunct to support standardized periodontal diagnosis, risk assessment, and recall planning, while comprehensive clinical examination and professional judgment remain essential.