Model-Based Orthodontic Assessments for Dental Panoramic Radiographs

Area of Science:

• Dentistry
• Medical Imaging
• Computer-Aided Diagnosis

Background:

• Accurate orthodontic evaluation is crucial for successful treatment outcomes.
• Current methods rely on manual parameter assessment, which can be subjective.
• Objective and reliable assessment tools are needed to support clinical decision-making.

Purpose of the Study:

• To propose and validate a novel automated method for orthodontic parameter assessment using dental panoramic radiographs.
• To enhance the accuracy and consistency of orthodontic evaluations.
• To provide dentists with a reliable tool for objective treatment assessment.

Main Methods:

• The proposed method involves four stages: image preprocessing, model training, automatic tooth segmentation, and orthodontic parameter assessment.
• Image preprocessing includes normalization and enhancement.
• Model training encompasses shape, image, energy function, and weight training, followed by energy-minimized tooth contour segmentation.

Main Results:

• The automated method demonstrated high accuracy, with average absolute distance, Dice similarity coefficient, and qualitative scores within specific ranges (4.17-6.03, 0.87-0.90, 2.58-3.12, respectively).
• The system's orthodontic assessments closely aligned with evaluations performed by experienced orthodontists.
• The method achieved accurate and consistent measurements.

Conclusions:

• The developed automated method effectively assists dentists in obtaining objective orthodontic treatment evaluations.
• The technique offers a reliable and consistent approach to measuring orthodontic parameters from panoramic radiographs.
• This innovation has the potential to improve the precision of orthodontic diagnosis and treatment planning.