3D generation of dental crown bottoms using context learning

Abstract

The generation of valid and realistic dental crown bottoms plays a central role in dentistry, as dental crown bottoms are the first point of contact between a tooth preparation and its crown. Every tooth is different, and the retention of the crown bottom heavily depends on how well it fits the preparation while conserving essential properties for ceramic adhesion and smoothness. From this, the generation of the crown bottom becomes a difficult task that only qualified individuals such as dental technicians can complete. Standard geometric modelling techniques such as computer-aided design (CAD) software programs have since been used for this purpose, providing a reliable basis for the generation of dental crown bottoms. Conversely, recent improvements in deep learning have presented new avenues in shape generation tasks that allow for personalized shapes to be created in a short period of time based on learned context. Starting from a set of preparation shapes, this project seeks to compare the efficacy of automatic geometric techniques to deep learning methods in the framework of dental crown bottom shape generation. Results show that deep learning methods such as GANs demand no human manipulation and provide similar visual results to the geometric model on unseen cases in an unsupervised manner. Our code is available at https://github.com/ImaneChafi/C.B.GEN and https://github.com/ImaneChafi/Prep-GAN

1. INTRODUCTION

A cavity is a serious infection of the teeth that must be treated quickly. Dental preparations remove this infection

and prepare the tooth to receive a crown. A crown bottom is the first point of contact of the new crown to

be put atop the prepared teeth. Due to the high degree of uniqueness in human teeth, poor crown bottom

design can lead to loose crowns, increasing the risk of cavity formation.1 Current methods for crown bottom

generation involve human manipulation by adding a cement gap and spacing for the margin line extension using

computer-aided design software (CAD). This takes time and expertise, and although current dentistry software

using computer aided design provide sliders and manual segmentation to generate the crown bottom from the

preparation, it demands human expert manipulation and lacks automation. To solve this problem, artificial

intelligence can help us generate dental crown bottoms automatically, thus alleviating the workload on dental

technicians and allowing more time for the design of crown shells. One of the many challenges in the case of

crown bottom generation using machine learning is its consistency, due to the unique nature of human teeth. Our

goal is to investigate and compare the use of automated geometric solutions, as well as deep learning solutions

that will allow for crown bottoms to be created in an automated manner, which in turn ameliorates the crown

generation pipeline. In addition, the proposed method will be compared to an automated geometric solution.

1.1 Related work

A crown bottom involves a set of geometric constraints to be followed to be as close as possible to the prepared

teeth, while still allowing enough space for cement to merge the crown and the preparation.2 A crown is comprised

of a shell, a margin line as the delimitation between the preparation and the crown and a crown bottom as the

shape that connects the preparation and the new crown. According to a paper by ELOS Medtech,3 the cement

gap should be in the range of microns for maximum crown retention. The cement gap closer to the margin line

should be near 0 microns. At 35 microns from the margin line, the crown bottom should be smoothed-out to

remove possible misshapes from undercuts.

Current research on crown generation456compared the performance of automatic CAD designs for crown

generation to dental technician’s work, and concluded that the performance is similar to experts. Research

following deep learning in dentistry such as this paper7 proposes the use of a two-stage generative adversarial

network based on the use of a conditional Generative Adversial Network (DCPR-GAN) that learns feature

correspondences between a crown and the previous teeth’s defects to solve occlusion issues,7 but do not use

machine learning methods for their crown bottom (adhesive layer). This article by Lessard et al.8 tests the

adoption of a GAN for the generation of a dental crown. The authors show that GANs provide the ability

to generate dental crowns through point completion, but that there exists difficulties in noise appearing if the

preparation is missing before crown generation. One of the main limitations of this model is the lack of high

precision detailing. Current state-of-the-art models for shape generation following anatomic shapes include

ShapeGF,9 LION10 and SP-GAN.11 This last article’s model follows a parts-aware deformation and generation,

through the use of a sphere global prior and a gaussian local prior, feeding a prior latent matrix for dense

correspondence. Using the well-known dataset ShapeNet,12 SP-GAN achieves better qualitative results, more

details and less noise than tree-GAN,13 ShapeGF9 and PointFlow.14 As the crown bottom is attached to the

shell, the need for a comprehensive model for its generation is an important step in the creation of the crown

shell. To the best of our knowledge, the process of automating the crown bottom shape using deep learning

and comparing it to current methods of CAD-based generation has yet to be studied and is important in future

crown generation steps, which showcases the critical nature of our research objective.

1.2 New work to be presented

The practicality of this field has yet to be studied in relation to how precisely a dental technician creates a crown

bottom, and whether artificial intelligence can help in the generation. We propose: 1) a novel fully-automatic

geometric method for full crown-bottom generation from a preparation shape; 2) an original unsupervised generation

from a GAN network machine learning method inspired by SP-GAN,11 specific to crown bottoms and a

comparison to the geometric method; 3) our code for both the geometric method and the deep learning method

is available online. We have also made our outputs for the GAN-based generated model available on our github.

You may read the full publication here or download the pdf version.

Dental Restoration using a Multi-Resolution Deep Learning Approach

Dental Restoration using a Multi-Resolution Deep Learning Approach

Published in SPIE. Digital Library, here’s an article published by our team explaining how our work in dental AI, through semi-supervised segmentation, improves the generalization of currently available deep learning segmentation model on 3D dental arches by introducing a new loss function to leverage unlabeled available data.

Dental offices tackle thousands of dental reconstructions every year. Complexity and abnormalities in dentition make segmentation of an optical scan a challenging manual task that takes 45 minutes on average. The present work improves the generalization of currently available deep learning segmentation model on 3D dental arches by introducing a new loss function to leverage unlabeled available data. The semi-supervised segmentation network is trained using a joint loss that combines a supervised loss of annotated input and a self-supervised loss of non-labeled input. Our results showed that combining self-supervised and supervised learning improved the segmentation score by 13 % compared with purely supervised learning for the same amount of labeled data. It is concluded that combining representations obtained from self-supervised learning with supervised learning improves the generalization of the 3D tooth segmentation model in the case of few available labeled data.

Ammar Alsheghri,  Farnoosh Ghadiri, Ying Zhang,  Olivier Lessard, Julia Keren,  Farida Cheriet, François Guibault

You may read the full publication here.

Here’s a presentation video by our team’s Ammar Alsheghri

Semi-supervised segmentation of tooth from 3D scanned dental arches

Published in SPIE. Digital Library, here’s an article published by our team explaining how our work in dental AI, through semi-supervised segmentation, improves the generalization of currently available deep learning segmentation model on 3D dental arches by introducing a new loss function to leverage unlabeled available data.

Dental offices tackle thousands of dental reconstructions every year. Complexity and abnormalities in dentition make segmentation of an optical scan a challenging manual task that takes 45 minutes on average. The present work improves the generalization of currently available deep learning segmentation model on 3D dental arches by introducing a new loss function to leverage unlabeled available data. The semi-supervised segmentation network is trained using a joint loss that combines a supervised loss of annotated input and a self-supervised loss of non-labeled input. Our results showed that combining self-supervised and supervised learning improved the segmentation score by 13 % compared with purely supervised learning for the same amount of labeled data. It is concluded that combining representations obtained from self-supervised learning with supervised learning improves the generalization of the 3D tooth segmentation model in the case of few available labeled data.

Ammar Alsheghri,  Farnoosh Ghadiri, Ying Zhang,  Olivier Lessard, Julia Keren,  Farida Cheriet, François Guibault

You may read the full publication here.

Here’s a full description by our team’s Ammar Alsheghri

Improving the quality of dental crown using a transformer-based method

Here’s a paper on SPIE. Medical Library published by our team.

Designing a synthetic crown is a time-consuming, inconsistent, and labor-intensive process. In this work, we present a fully automatic method that not only learns human design dental crowns, but also improves the consistency, functionality, and esthetic of the crowns. Following success in point cloud completion using the transformer-based network, we tackle the problem of the crown generation as a point-cloud completion around a prepared tooth.

You may read the full publication here.

3D generation of dental crown bottoms using context learning

Here’s a publication in SPIE. Medical Imaging, presented by one of our researchers, Imane Chafi, who is a PhD student at Polytechnique Montreal currently developing ML models for dental restoration and evaluation. Her research interests are in medical imaging, 3D shape generation and multimodal matching.

The generation of valid and realistic dental crown bottoms plays a central role in dentistry, as dental crown bottoms are the first point of contact between a tooth preparation and its crown. Every tooth is different, and the retention of the crown bottom heavily depends on how well it fits the preparation while conserving essential properties for ceramic adhesion and smoothness. From this, the generation of the crown bottom becomes a difficult task that only qualified individuals such as dental technicians can complete. Standard geometric modelling techniques such as Computer-Aided Design (CAD) software programs have since been used for this purpose, providing a reliable basis for the generation of dental crown bottoms.

Imane Chafi, François Guibault, Julia Keren, Ying Zhang and Farida Cheriet

You may read the full publication here.

MC-Net: Mesh Completion for Dental Scans

Abstract — Dental centers need to design hundreds of dental crowns per year using computer assisted design (CAD) software. Typically, technicians manually modify a template shape to create crowns. That process requires a lot of time and experience to be done correctly, which leads to great variability in quality. In recent years, many deep
learning methods have been proposed to do point cloud completion by predicting only the missing region. Although these methods are potentially applicable to the task of dental crown design, most of them fail to generate smooth point clouds, which is critical for surface reconstruction. In this paper, we propose an end-to-end approach called MCNet for automatic mesh completion of dental scans. Using an input point cloud sampled at multiple resolutions and a template shape for the type of tooth to generate, MC-Net extracts features to guide a mesh deformation. The mesh generation follows a coarse-to fine strategy and uses a mesh-related loss function to make the procedure stable. Our model can generate visually correct and accurate surfaces of the missing regions.

Our dental AI project described in depth by our team of researchers and engineers. In this publication, you can learn more about the design and generation of perfect dental crowns using artificial intelligence and machine learning.

In dental centers, technicians need to manually design hundreds of crowns per year. Crown generation implies designing an external surface that is visible once the crown is installed on the tooth preparation. That design is challenging even with today’s computer assisted design (CAD) software. The technician must design a tooth shape with a complex morphology and patient-specific characteristics. Usually, the dentist starts by making a preparation with the damaged tooth. The preparation serves as a foundation on which the dental crown will be installed. Secondly, the dentist scans the prepared tooth and the surrounding teeth with an intra-oral scanner to get a 3D representation. Finally, the technician uses that surrounding information to design a patient-specific crown.

Olivier Lessard, Member, IEEE, François Guibault, Member, IEEE, Julia Keren and Farida Cheriet, Senior Member, IEEE

You may read the full publication here.

from-mesh-completion-to-AI-designed-crown

From Mesh Completion to AI Designed Crown

Publication in Springer Link describing our project and its importance in the dental profession for the generation of the perfect dental crown using artificial intelligence.

You may read the abstract and the full publication on Springer Link.

Improving-the-quality-of-dental-crown-using-a-transformer-based-method

Improving the quality of dental crown using a transformer-based method

Article published by our team in SPIE. Digital Library explaining our project’s contribution to the design and creation of synthetic dental crowns.

In this publication, we present a fully automatic method that not only learns human design dental crowns, but also improves the consistency, functionality, and the aesthetic aspect of the dental crowns.

Read the abstract and access the full article here.

teeth made with dental AI

Making Teeth Using Dental AI

teeth made with dental AI

Here’s an article describing the dental AI project of Intellident, published in the reputable Le Devoir newspaper. The article (in French), references an interview with Professor Francois Guildbault, the lead professor of our project.

In the interview, Prof. Guildbault explains how using machine learning and artificial intelligence, we aim to simplify and perfect the creation of dental restorations (implants, crowns, etc.). Most importantly, he highlights how this will help dental laboratories everywhere to create the perfect restoration quickly and perfectly.

Une équipe de chercheurs de Polytechnique Montréal utilise l’intelligence artificielle pour réduire les marges d’erreurs dans la confection de couronnes dentaires.

L’hypertrucage au service de la dentisterie, telle est la devise de François Guibault, professeur titulaire au Département de génie informatique et génie logiciel de Polytechnique Montréal. Depuis 18 mois, le chercheur utilise la même technologie d’intelligence artificielle (IA) qui sert à falsifier des vidéos pour fabriquer des couronnes dentaires.

You may read the full article on Le Devoir’s website.

About Dental AI

In recent years, the integration of Artificial Intelligence (AI) in dentistry has been a game-changer, revolutionizing the way oral healthcare is delivered. Dental AI, a cutting-edge technology, is proving to be a valuable ally for both dentists and patients.

One of the primary benefits of Dental AI is its ability to enhance diagnostic accuracy. Advanced imaging algorithms analyze dental scans with precision, aiding in the early detection of dental issues such as cavities, gum disease, and abnormalities. This not only ensures timely intervention but also contributes to more effective treatment planning.

Moreover, Dental AI streamlines administrative tasks, allowing dental professionals to focus more on patient care. Automated appointment scheduling, billing processes, and patient reminders are just a few examples of how AI is optimizing the workflow in dental practices, leading to improved efficiency and patient satisfaction.

The utilization of AI in treatment planning is another noteworthy aspect and is mainly the area of the project at Intellident AI. By analyzing patient data and historical records, Dental AI assists in personalized treatment plans, ensuring optimal outcomes. This tailored approach enhances the overall patient experience and contributes to long-term oral health. With the help of the software the Intellident AI team is working on, dental laboratories will be able to use this data to automate the creation of perfectly matching dental crowns and other restorations.

In conclusion, Dental AI is reshaping the landscape of oral healthcare, bringing forth unprecedented efficiency and precision. As the technology continues to evolve, dentists and patients alike can expect even more advanced solutions, ultimately leading to a brighter, healthier future for dental care. Embrace the future of dentistry with Dental AI – where innovation meets optimal oral health.

Learn more about our vision and how we aim to bring AI to the dental profession.