Peer-influenced content. Sources you trust. No registration required. This is HCN.

Journal of Dental ResearchMental Health and Oral Health in a Nationally Representative Cohort

How is machine learning revolutionizing dental prognosis models and what implications does it have for your practice?


Emerging research from southern Brazil showcases the power of machine learning (ML) in developing reliable dental caries prognosis models for primary and permanent teeth. By employing ML algorithms, the study proposes a new way of predicting dental caries, leveraging various factors identified during early childhood.

Key Points:
  • The study developed caries prognosis models for primary and permanent teeth using machine learning, based on predictors identified in early childhood.
  • A 10-year prospective cohort study was conducted in southern Brazil involving children aged 1 to 5 years, examined first in 2010 and reassessed in 2012 and 2020.
  • Machine learning algorithms used include decision tree, random forest, and extreme gradient boosting (XGBoost) alongside logistic regression.
  • The models achieved an area under the receiver operating characteristic curve (AUC) above 0.70 when predicting caries in primary teeth after a 2-year follow-up, with baseline caries severity being the strongest predictor.
  • After a 10-year follow-up, the SHAP algorithm based on XGBoost achieved an AUC higher than 0.70, indicating caries experience, nonuse of fluoridated toothpaste, parental education, frequency of sugar consumption, frequency of visits to relatives, and parents’ perception of their children’s oral health as top predictors for caries in permanent teeth.
Additional Points:
  • From the initial 639 children, 467 (73.3%) and 428 (66.9%) were reassessed in 2012 and 2020 respectively.
  • Dental caries were assessed using the International Caries Detection and Assessment System (ICDAS) criteria.
  • Demographic, socioeconomic, psychosocial, behavioral, and clinical factors were also considered in the analysis.
Conclusion:
  • The study concludes that machine learning presents significant potential in determining caries development in both primary and permanent teeth, utilizing easily collectable predictors from early childhood.

Further Reading

Did You Know?
According to the World Health Organization, dental caries remains a major oral health problem in most industrialized countries, affecting 60-90% of schoolchildren.

The Healthcare Communications Network is owned and operated by IQVIA Inc.

Click below to leave this site and continue to IQVIA’s Privacy Choices form