Assessment of artificial intelligence in detecting errors on panoramic radiographs

View Full Paper

Ask AI

Bookmark

Share

Key Points

The study aims to evaluate artificial intelligence's ability to detect common errors in panoramic radiographs.
Utilized a dataset of 2,888 anonymized panoramic radiographs from various dental imaging units.
Classified images into three categories: uneven magnification, tongue space, and normal.
Developed and evaluated three deep learning architectures: Attention Dental X-ray, ResNet50, and MobileNet.
Assessed model performance using accuracy, precision, sensitivity, area under the curve, and F1 score with 5-fold cross-validation.
Applied statistical analysis using Python libraries and assessed model comparison using McNemar's test.
The Attention Dental X-ray model exhibited the best performance among all models evaluated.
Achieved an average accuracy of 82.7% in classifying tongue space errors.
ResNet50 and MobileNet models achieved accuracies of 56.9% and 64.1%, respectively.

We only have this paper's title and publication details.

Retrospective study assesses AI's effectiveness in detecting errors in panoramic radiographs, suggesting implications for dental quality assurance.

No takes yet. Share an insight, caveat, or question.

Salam et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark

Share