Integrating AI-Driven Radiographic Analytics with Salivary Biomarker Profiling for the Precision Diagnosis and Prognostic Prediction of Stage IV Periodontitis
- May 13
- 2 min read
DOI: https://doi.org/10.66715/jsccr/2026v3.i3.16 | Original Research | 2026 | Volume 3 | Issue 3 | Page 1-6
Dr. Shruti Wankhade, Reader, Department of Periodontology, VYWS Dental College and Hospital Amravati. (Corresponding Author)
Abstract
Background: Stage IV periodontitis is characterized by severe bone loss, tooth mobility, and a high risk of tooth loss, demanding a transition from conventional clinical staging to precision diagnostics. While radiographic assessments evaluate past structural damage, they lack real-time insights into active disease progression. This study evaluates the diagnostic accuracy and prognostic value of integrating Artificial Intelligence (AI)-driven radiographic analytics with salivary biomarker profiling for Stage IV periodontitis.
Methods: A cross-sectional study was conducted involving 120 subjects diagnosed with Stage IV periodontitis. Digital panoramic and intraoral periapical radiographs were analyzed using a trained Convolutional Neural Network (CNN) to quantify alveolar bone loss percentage and pattern. Concurrently, saliva samples were analyzed using enzyme-linked immunosorbent assays (ELISA) to profile inflammatory biomarkers, specifically Matrix Metalloproteinase-8 (MMP-8), Interleukin-1 beta, and Tumour Necrosis Factor-alpha. A machine learning algorithm merged radiographic and biomolecular data to predict disease stability over a 12-month follow-up.
Results: The AI model demonstrated high precision (94.2/%) in mapping complex bone defects compared to manual clinician assessment. High salivary concentrations of MMP-8 and IL-1 / beta significantly correlated with rapid, ongoing bone destruction (p < 0.001). The integrated multimodal AI-biomarker model achieved an Area Under the Curve (AUC) of 0.91 for predicting future tooth loss and disease progression, vastly outperforming standalone clinical staging (0.74 AUC).
Conclusion: Integrating AI radiographic analytics with salivary biomarker profiling offers a robust, objective, and minimally invasive approach to precision periodontics. This multimodal framework enhances diagnostic accuracy and provides reliable prognostic predictions, enabling personalized treatment pathways for Stage IV periodontitis patients.
Keywords: Stage IV Periodontitis, Artificial Intelligence, Deep Learning, Salivary Biomarkers, MMP-8, Precision Diagnostics, Prognosis.