Prediction of Lower Third Molar Eruption in Panoramic Radiography Using Artificial Intelligence (AI): PDApp

Area of Science:

• Dentistry
• Oral Surgery
• Medical Imaging
• Artificial Intelligence

Background:

• Predicting third molar eruption preoperatively is a significant challenge in dentistry and oral surgery.
• Current methods often lack precision, necessitating improved diagnostic tools.

Purpose of the Study:

• To develop and validate an AI-based tool for the preoperative prediction of third molar eruption.
• To assess the performance of machine learning algorithms in analyzing panoramic radiography for eruption diagnosis.

Main Methods:

• Machine learning (ML) algorithms were employed to analyze M3 panoramic radiological images.
• The study utilized retrospective data from patients with confirmed mandibular third molar retention.
• A dataset of 383 panoramic radiographies was used to train and validate the PDApp software.

Main Results:

• The PDApp software demonstrated high performance in differentiating between third molar eruption and retention.
• Clinical exploration confirmed the eruption diagnoses used for validation.
• The AI tool showed promising results in preoperative diagnostic accuracy.

Conclusions:

• The PDApp software, utilizing AI and ML, offers a high-performance solution for predicting mandibular third molar eruption.
• This AI-driven approach can aid in preoperative diagnosis, potentially improving treatment planning in oral surgery.