Artificial Intelligence Automatically Locates Mandibular Canals

Dentistry Today


Dentists need to know the exact location of the mandibular canal, where the alveolar nerve resides, when they’re planning dental implant procedures. Typically, dentists and radiologists use X-rays or CT scans to define the location of these canals, which can be laborious and time-consuming. Artificial intelligence, however, could make this work much easier. 

The Finnish Center for Artificial Intelligence, Tampere University Hospital, Planmeca, and the Alan Turing Institute have collaborated on a model that accurately and automatically shows the exact location of mandibular canals. Using deep neural networks, the researchers trained the model by using a dataset of 3-D CBCT scans. 

The model is based on a fully convolutional architecture to make it as fast and data-efficient as possible. Based on the research results, this type of deep learning model can very accurately localize mandibular canals. It surpasses the statistical shape models, which thus far have been the best automated method for localizing mandibular canals, the researchers said. 

In simple cases when the patient doesn’t have any special conditions such as osteoporosis, the model is as accurate as a human specialist. Most patients who visit a dentist fall into this category, the researchers said.

“In more complex cases, one may need to adjust the estimate, so we are not yet talking about a fully standalone system,” said Joel Jaskari, doctoral candidate at the Aalto University School of Science and first author of the study.

Artificial intelligence also enables the machine to perform the job equally fast and accurately every time, the researchers said. 

“The aim of this research work is not, however, to replace radiologists but to make their job faster and more efficient so that they will have time to focus on the most complex cases,” said Kimmo Kaski, senior advisor in computational science at Aalto University.

Planmeca is currently integrating the presented model into its dedicated software to be used with its 3-D tomography equipment.

The study, “Deep Learning Method for Mandibular Canal Segmentation in Dental Cone Beam Computed Tomography Volumes,” was published by Scientific Reports.

Related Articles

Artificial Intelligence Goes Back to the Future

Artificial Intelligence: The Future of Dentistry

Artificial Intelligence Automatically Calculates Orthodontic Measurements