According to a study showcased at EuroPerio10, the world’s leading congress in periodontology and implant dentistry arranged by the European Federation of Periodontology (EFP), a deep learning algorithm has positively detected periodontal disease from two-dimensional (2D) bitewing radiographs.
EuroPerio10, the world`s congress in periodontology and implant dentistry, started yesterday in Copenhagen. Image Credit: European Federation of Periodontology (EFP)
Our study shows the potential for artificial intelligence (AI) to automatically identify periodontal pathologies that might otherwise be missed. This could reduce radiation exposure by avoiding repeat assessments, prevent the silent progression of periodontal disease, and enable earlier treatment.
Dr. Burak Yavuz, Study Author, Eskisehir Osmangazi University, Turkey
Earlier studies have analyzed the usage of AI to identify root fractures, caries, and apical lesions but there is minimal research in the domain of periodontology. This research assessed the capacity of deep learning, a category of AI, to establish the status of periodontitis in bitewing radiographs.
The research utilized 434 bitewing radiographs from patients having periodontitis. Image processing was carried out with u-net architecture, a convolutional neural network used to rapidly and accurately segment images. An expert physician also assessed the images using the segmentation technique.
Evaluations included total alveolar bone loss around the upper and lower teeth, vertical bone loss, horizontal bone loss, furcation defects, and calculus around mandibular and maxillary teeth.
The neural network detected 2,215 cases of horizontal bone loss, 108 furcation defects, 859 cases of alveolar bone loss, 340 cases of vertical bone loss, and 508 cases of dental calculus. The algorithm’s success at detecting defects was compared against the assessment of a physician and reported as sensitivity, precision, and F1 score, which is the weighted average of sensitivity and precision. For the sensitivity, precision, and F1 score, 1 is considered the best value and 0 is considered the worst.
The sensitivity, precision, and F1 score results were 1, 0.94, and 0.96, respectively for total alveolar bone loss. The corresponding values for horizontal bone loss were 1, 0.92, and 0.95, respectively while AI could not detect vertical bone loss. With regards to dental calculus, the sensitivity, precision, and F1 score results were 1.0, 0.7, and 0.82, respectively and for furcation defects, the corresponding values were 0.62, 0.71, and 0.66, respectively.
Our study illustrates that AI is able to pick up many types of defects from 2D images which could aid in the diagnosis of periodontitis. More comprehensive studies are required on larger data sets to increase the success of the models and extend their use to 3D radiographs.
Dr. Burak Yavuz, Study Author, Eskisehir Osmangazi University, Turkey
Dr. Burak Yavuz concluded: “This study provides a glimpse into the future of dentistry, where AI automatically evaluates images and assists dental professionals to diagnose and treat disease earlier.”