Exploring the accuracy of tooth loss prediction between a clinical periodontal prognostic system and a machine learning prognostic model

• Published in: Journal of clinical periodontology Vol. 51; no. 10; pp. 1333 - 1341
• Main Authors: Santamaria, Pasquale, Troiano, Giuseppe, Serroni, Matteo, Araùjo, Tiago G., Ravidà, Andrea, Nibali, Luigi
• Format: Journal Article
• Language: English
• Published: Oxford, UK Blackwell Publishing Ltd 01-10-2024
• Subjects: Artificial intelligence; Datasets; diagnosis; Neural networks; periodontal diseases; Periodontitis; Prediction models; Prognosis; Regression analysis; Teeth; tooth loss; United States > US; Pittsburgh Pennsylvania; diagnosis; periodontal diseases; prognosis; tooth loss
• ISSN: 0303-6979; 1600-051X; 1600-051X
• DOI: 10.1111/jcpe.14023

Aim
The aim of this analysis was to compare a clinical periodontal prognostic system and a developed and externally validated artificial intelligence (AI)‐based model for the prediction of tooth loss in periodontitis patients under supportive periodontal care (SPC) for 10 years.
Materials and Methods
Clinical and radiographic parameters were analysed to assign tooth prognosis with a tooth prognostic system (TPS) by two calibrated examiners from different clinical centres (London and Pittsburgh). The prediction model was developed on the London dataset. A logistic regression model (LR) and a neural network model (NN) were developed to analyse the data. These models were externally validated on the Pittsburgh dataset. The primary outcome was 10‐year tooth loss in teeth assigned with ‘unfavourable’ prognosis.
Results
A total of 1626 teeth in 69 patients were included in the London cohort (development cohort), while 2792 teeth in 116 patients were included in the Pittsburgh cohort (external validated dataset). While the TPS in the validation cohort exhibited high specificity (99.96%), moderate positive predictive value (PPV = 50.0%) and very low sensitivity (0.85%), the AI‐based model showed moderate specificity (NN = 52.26%, LR = 67.59%), high sensitivity (NN = 98.29%, LR = 91.45%), and high PPV (NN = 89.1%, LR = 88.6%).
Conclusions
AI‐based models showed comparable results with the clinical prediction model, with a better performance in specific prognostic risk categories, confirming AI prediction model as a promising tool for the prediction of tooth loss.