Tackling gaps in peri-implantitis detection with deep-learning model

A deep-learning model could help physicians identify peri-implantitis on orthopantomographs.

In a study published in the Journal of Evidence-Based Dental Practice, researchers used a deep-learning-based approach to examine 7,696 orthopantomographs for peri-implantitis. Following segmentation and conversion of DICOM-formatted images into PNG files, they then applied a convolutional neural network to distinguish healthy implants and those with peri-implantitis.

The researchers found that the model demonstrated an accuracy of 0.999, Dice Similarity Coefficient of 0.986 and Intersection over Union of 0.974. Further, among the 3,693 implants included in the study, 638 of them had peri-implantitis — 576 of which were clinically identified by the deep-learning model as having peri-implantitis.

The findings could help improve the diagnostic accuracy and treatment planning of peri-implantitis.

Read more: Journal of Evidence-Based Dental Practice

The article presented here is intended to inform you about the broader media perspective on dentistry, regardless of its alignment with the ADA's stance. It is important to note that publication of an article does not imply the ADA's endorsement, agreement, or promotion of its content.