Predicting periodontitis with machine-learning model

A novel machine-learning model could effectively predict the risk of developing periodontal disease.

In a study published in the Journal of Dentistry, researchers created a machine-learning model capable of identifying periodontitis risk by assessing nonimage electronic dental records. They trained the model with the dental records of U.S. patients participating in the BigMouth repository and subsequently conducted analyses of the model’s accuracy, sensitivity, specificity and area under the curve.

The researchers noted that compared with four other machine-learning models, the Random Forest model exhibited the greatest sensitivity and area under the curve. They found that factors such as bleeding proportion, age, number of visits, prior preventive treatment, smoking status and drug use were the most predictive of periodontitis. Following validation, the novel machine-learning model was found to accurately predict 91% of the cases but overestimated the risk among those who hadn’t developed the disease.

The findings indicated that the machine-learning model could improve the early detection of periodontitis and allow for the earlier implementation of preventive therapy.

Read more: Journal of Dentistry

The article presented here is intended to inform you about the broader media perspective on dentistry, regardless of its alignment with the ADA's stance. It is important to note that publication of an article does not imply the ADA's endorsement, agreement, or promotion of its content.