Clinical Decision Support System for Geriatric Dental Treatment Using a Bayesian Network and a Convolutional Neural...

Area of Science:

• Geriatric Dentistry
• Artificial Intelligence in Healthcare
• Clinical Decision Support Systems

Background:

• Therapeutic planning in geriatric dentistry requires integrating oral health status with factors like comorbidities and functional dependency.
• Existing methods may not comprehensively address the multifaceted needs of geriatric dental patients.

Purpose of the Study:

• To evaluate the performance of a novel clinical decision support system (CDSS) for formulating therapeutic plans in geriatric dentistry.
• To assess the system's ability to incorporate diverse patient data for treatment recommendations.

Main Methods:

• A Bayesian network (BN) model was developed, integrating clinical knowledge and data.
• The faster R-CNN algorithm was employed for oral health status detection within the BN.
• Retrospective data from 400 geriatric dental patients were analyzed.

Main Results:

• The system achieved an F1-score of 89.31% for detecting periodontally compromised teeth.
• The BN model demonstrated high accuracy in recommending therapeutic plans (AUC = 0.902).
• The CDSS showed strong agreement with geriatric dentistry experts (kappa = 0.905).

Conclusions:

• This study represents a foundational phase in developing an AI-driven CDSS for geriatric dental treatment.
• The proposed system is anticipated to offer general practitioners expert-level decision support for geriatric dental care.
• Integration into clinical workflows is expected to enhance the quality of care for elderly dental patients.