Machine Learning in Predicting Tooth Loss: A Systematic Review and Risk of Bias Assessment

Abstract

Predicting tooth loss is a persistent clinical challenge in the 21st century. While an emerging field in dentistry, computational solutions that employ machine learning are promising for enhancing clinical outcomes, including the chairside prognostication of tooth loss. We aimed to evaluate the risk of bias in prognostic prediction models of tooth loss that use machine learning. To do this, literature was searched in two electronic databases (MEDLINE via PubMed; Google Scholar) for studies that reported the accuracy or area under the curve (AUC) of prediction models. AUC measures the entire two-dimensional area underneath the entire receiver operating characteristic (ROC) curves. AUC provides an aggregate measure of performance across all possible classification thresholds. Although both development and validation were included in this review, studies that did not assess the accuracy or validation of boosting models (AdaBoosting, Gradient-boosting decision tree, XGBoost, LightGBM, CatBoost) were excluded. Five studies met criteria for inclusion and revealed high accuracy; however, models displayed a high risk of bias. Importantly, patient-level assessments combined with socioeconomic predictors performed better than clinical predictors alone. While there are current limitations, machine-learning-assisted models for tooth loss may enhance prognostication accuracy in combination with clinical and patient metadata in the future.

Systematic Review of Prognosis Models in Predicting Tooth Loss in Periodontitis

“…The strengths of this study include the large number of studies included in the analysis and the focus on model performance, facilitating a clear-eyed assessment of the variation in model performance within and across models, as well as helping to extrapolate the performance of future models. Prior reviews on related areas focused much more narrowly on population-based samples (Du et al 2018) or machine learning (Hasuike et al 2022). This is also the first review to use the CHARMS checklist in the field of periodontology.…”

Development and validation of intraoral periapical radiography-based machine learning model for periodontal defect diagnosis

“…In recent years, innovative computer techniques, especially artificial intelligence (AI), have begun to be used in many areas of dentistry and are helping increase treatment and diagnostic demands. 12–15 The concept of “artificial intelligence” denotes the ability of machines to perform tasks normally completed by humans. 16 Different techniques are used in AI like machine learning and deep learning algorithm.…”

Identifying predictors of tooth loss using a rule‐based machine learning approach: A retrospective study at university‐setting clinics

“…In recent years, machine learning algorithms and deep learning models have been widely used to build diagnosis and prediction tools for dental diseases 13–18 . However, deep learning models are complicated black box models that usually fall short of interpretability.…”

“…In recent years, machine learning algorithms and deep learning models have been widely used to build diagnosis and prediction tools for dental diseases. [13][14][15][16][17][18] However, deep learning models are complicated black box models that usually fall short of interpretability. Machine learning models, especially tree-based methods, provide great potential for accurate prediction while preserving model interpretability.…”