Comparative Accuracy of AI Models and Clinical Assessment for Dental Plaque Detection in Children

Study Title:

Accuracy of Dental Plaque Detection from Intraoral Images Using Different Artificial Intelligence Models Versus Clinical Assessment Among a Group of Children: A Diagnostic Accuracy Study

Study Overview:

This observational diagnostic accuracy study is designed to evaluate the performance of multiple artificial intelligence (AI) models in detecting dental plaque from intraoral images, compared to traditional clinical assessments conducted by qualified dentists. The primary focus is on pediatric patients, as early detection and management of dental plaque are crucial for maintaining oral health in children.

Background and Rationale:

Dental plaque is a biofilm that forms on teeth and can lead to caries and periodontal disease if not properly managed. Traditional methods of plaque detection rely on visual assessments by dental professionals, which can be subjective and may vary in accuracy. Recent advancements in AI and image processing present an opportunity to enhance the detection and quantification of dental plaque through intraoral images, potentially providing a more objective and efficient assessment tool.

Objectives:

To compare the accuracy of AI models in detecting dental plaque against clinical assessments.

To determine the sensitivity, specificity, and overall diagnostic performance of the AI technologies.

To analyze the potential for AI models to be integrated into routine dental examinations for pediatric patients.

Methodology:

Participants: A sample of pediatric patients will be recruited, ensuring a diverse representation of various demographics and dental health statuses.

Image Acquisition: Intraoral images will be captured using standardized imaging protocols to ensure consistency. High-resolution images will be obtained under controlled conditions to minimize variability.

AI Models: Various AI algorithms, including convolutional neural networks (CNNs) and deep learning techniques, will be trained using a dataset of annotated intraoral images. These models will be evaluated based on their ability to identify and quantify dental plaque.

Clinical Assessment: Trained dentists will perform clinical examinations using standard plaque indices to assess the presence and severity of dental plaque in the same cohort of children.

Data Analysis: Statistical methods will be employed to compare the diagnostic accuracy of AI models with clinical assessments, including calculations of sensitivity, specificity, positive predictive value, and negative predictive value.

Expected Outcomes:

The study aims to elucidate the role of AI in enhancing the detection of dental plaque in children, potentially leading to improved preventive care and treatment strategies. The findings may also contribute to the development of AI-assisted tools for dental practitioners.

Ethical Considerations:

This study will adhere to ethical guidelines, ensuring informed consent is obtained from legal guardians of pediatric participants. Approval from the relevant institutional review board (IRB) will be secured prior to the commencement of the study