Study Finds Machine Learning Models Can Predict Caries Risk - Scorecard - DentalSpire
Coming Soon: Introducing MDSpire News. Learn more
Conexiant’s news site is now MDSpire News. Learn more

Study Finds Machine Learning Models Can Predict Caries Risk

  • By

  • News Staff

  • November 9, 2026

Share

Clinical Scorecard: Study Finds Machine Learning Models Can Predict Caries Risk

At a Glance

CategoryDetail
ConditionDental caries presence and severity
Key MechanismsMachine learning models used demographic, socioeconomic, behavioral, and clinical data; wealth and household density were the leading structural predictors.
Target PopulationBrazilian survey participants aged 12, 15–19, and 35–44 years.
Care SettingPopulation-level oral health surveillance and public health risk stratification in urban Brazilian populations.

Key Highlights

  • Models predicted caries presence and severity, with performance varying by age group and stronger discrimination of caries presence among adults.
  • Explainability analysis found socioeconomic and structural factors outweighed individual clinical markers across all age groups.
  • Wealth and household density were the primary structural predictors of caries risk.
  • Across age groups, non-White participants had a higher caries burden; private dental insurance and recent dental visits were associated with lower caries prevalence.
  • Among 12-year-olds, untreated caries prevalence was 62.5% without piped water access and 35.5% with access.

Guideline-Based Recommendations

Diagnosis

  • The study assessed untreated caries presence and disease severity through clinical examinations.

Management

  • The authors describe AI-based risk stratification as a potential support for targeting public health programs and preventive care to specific groups.

Monitoring & Follow-up

  • The authors propose repeating the method with country-specific factors; a parallel analysis using US National Health and Nutrition Examination Survey data is in preparation.

Risks

  • The cross-sectional design does not support causal conclusions.
  • The study focused on urban populations and did not externally validate the models.
  • Findings may not translate directly to countries with different dental care access and socioeconomic conditions.

Patient & Prescribing Data

The analysis included 6,698 12-year-old children, 8,041 adolescents aged 15–19, and 8,987 adults aged 35–44 from Brazil's 2023 Oral Health Survey.

No treatment or prescribing regimen was evaluated. The study examined caries risk prediction using demographic, socioeconomic, behavioral, and clinical data.

Clinical Best Practices

  • Interpret the reported associations as population-level findings; the cross-sectional study cannot establish causation.
  • Consider the study's urban focus and lack of external validation when interpreting model performance or applying findings elsewhere.
  • The authors identify country-specific factors as relevant to developing population-tailored prevention strategies.

Related Resources & Content

Original Source(s)

Related Content