ARTIFICIAL INTELLIGENCE
Deep learning tools for dentomaxillofacial application
The rationale behind this topic:
The use of Artificial Intelligence methods in dental practice, such as deep learning, brings new perspectives for the diagnosis, classification and prediction of oral diseases, for treatment planning, and for the evaluation and prediction of outcomes. Deep convolutional neural networks hold promise to improve both quality and quantity of future image processing, and may help radiologists to analyze images that contain vast data information This project aims to evaluate the performance of combined deep convolutional neural networks for automatic tooth segmentation on dental panoramic radiographs (2D segmentation) and on CBCT scans (3D Segmentation).
contact:
Professor Reinhilde Jacobs
reinhilde.jacobs@uzleuven.be
Related projects
Project 1: Accurate and fast deep learning tool for tooth detection and segmentation on panoramic radiographs
Researchers: André Ferreira Leite, Adriaan Van Gerven, Holger Willems, Thomas Beznik, Pierre Lahoud, Hugo Gaêta-Araujo, Myrthel Vranckx, Reinhilde Jacobs
Published on:
Project 2: AI-driven molar angulation measurements to predict third molar eruption on panoramic radiographs
Researchers: Myrthel Vranckx, Adriaan Van Gerven, Holger Willems, Arne Vandemeulebroucke, André Ferreira Leite, Constantinus Politis, Reinhilde Jacobs
Published on:
Project 3: A novel artificial intelligence tool for accurate tooth segmentation on CBCT
Researchers: Pierre Lahoud, Mostafa EzEldeen, Thomas Beznik, Holger Willems, André Ferreira Leite, Adriaan Van Gerven, Reinhilde Jacobs
Published on:
Project 4: Automatic segmentation of the pharyngeal airway space with convolutional neural network
The rationale behind this topic:
Pharyngeal airway space assessment has been an area of interest for clinicians allowing efficient diagnosis, treatment planning and follow-up of patients with dento-skeletal deformities and obstructive sleep apnea which might influence the dimensions of upper airway space. Recently, deep learning convolutional neural networks (CNNs), has gained much attention in the dentomaxillofacial field for segmenting and evaluating the airway space. This project aims to propose and investigate the performance of a deep learning-based 3D CNN model for PAS segmentation from CT/CBCT images.
Researchers: Sohaib Shujaat, Omid Jazil, Holger Willems, Adriaan Van Gerven, Eman Shaheen, Constantinus Politis, Reinhilde Jacobs.
Published on:
Project 5: Development and validation of a novel artificial intelligence driven tool for accurate mandibular canal segmentation on CBCT
The objective of this study is the development and validation of a novel artificial intelligence driven tool for fast and accurate mandibular canal segmentation on cone beam computed tomography (CBCT).
Given the importance of adequate pre-operative mandibular canal assessment, Artificial Intelligence could help relieve practitioners from the delicate and time-consuming task of manually tracing and segmenting this structure, helping prevent per- and post-operative neurovascular complications.
Researchers: Pierre Lahoud, Siebe Diels, Liselot Niclaes, Stijn Van Aelst, Holger Willems, Adriaan Van Gerven, Marc Quirynen, Reinhilde Jacobs.
Published on:
Project 6: Influence of dental fillings on the performance of a novel artificial intelligence-driven tool for automatic tooth segmentation on CBCT images
Tooth segmentation is a challenging and time-consuming task, mainly in the presence of artifacts generated by dental filling material. The proposed AI-driven tool could offer a clinically acceptable approach for tooth segmentation, to be applied in the digital dental workflows considering its time efficiency and high accuracy even in the presence of artifacts generated from dental fillings.
Researchers: Rocharles Cavalcante Fontenele, Maurício do Nascimento Gerhardt, Jáder Camilo Pintom, Adriaan Van Gerven, Holger Willems, Reinhilde Jacobs, Deborah Queiroz Freitas
Published on:
Project 7: Layered deep learning for automatic mandibular segmentation in cone-beam computed tomography
The aim of this project was to develop and validate a layered deep learning algorithm that automatically creates 3D surface models of the human mandible from a CBCT scan. The hypothesis was that such an AI model could provide accurate 3D surface models of the mandible in a more reliable and time efficient way than the current clinical standard, being semi-automatic segmentation. At the end of this project, we established that a layered 3D U-Net architecture deep learning algorithm, with and without additional user refinements, improved time-efficiency, reduced operator error, and provided excellent accuracy when benchmarked against the clinical standard, being semi-automatic segmentation.
Researchers: Pieter-Jan Verhelst, Andreas Smolders, Thomas Beznik, Jeroen Meewis, Arne Vandemeulebroucke, Eman Shaheen, Adriaan Van Gerven, Holger Willems, Constantinus Politis, Reinhilde Jacobs
Published on:
Project 8: A novel deep learning system for multi-class tooth segmentation and classification on cone beam computed tomography. A validation study
Automatic tooth segmentation and classification from cone beam computed tomography (CBCT) have become an integral component of the digital dental workflows. Therefore, we developed and validated a deep learning approach for automatic tooth segmentation and classification from CBCT images with a dataset of 186 scans acquired with 2 different CBCT machines. The AI framework correctly segmented teeth with optimal precision (0.98±0.02) and 1800 times faster compared to an expert. Teeth classification also performed optimally with precision of 97.9%.
The proposed 3D U-Net based AI framework is an accurate and time-efficient deep learning system for automatic tooth segmentation and classification without expert refinement which indicates its potential future applications for diagnostics and treatment planning in the field of digital dentistry, while reducing clinical workload.
Researchers: Eman Shaheen, André Leite, Khalid Ayidh Alqahtani, Andreas Smolders, Adriaan Van Gerven, Holger Willems, Reinhilde Jacobs
Published on:
Project 9: Automated detection and labelling of teeth and small edentulous regions on cone-beam computed tomography using convolutional neural networks
This paper focuses on the accuracy of AI to identify and label teeth and missing teeth in small edentulous areas. Results showed excellent accuracy to perform both tasks in a fast manner. AI may represent a promising time-saving tool for radiological reporting, automated dental charting and surgical and treatment planning.
Researchers: Maurício do Nascimento Gerhardt, Rocharles Cavalcante Fontenele, André Ferreira Leite, Pierre Lahoud, Adriaan Van Gerven, Holger Williems, Andreas Smolders, Thomas Beznik, Reinhilde Jacobs
Published on:
Project 10: Convolutional neural network for automatic maxillary sinus segmentation on cone-beam computed tomographic images
An accurate three-dimensional (3D) segmentation of the maxillary sinus is crucial for multiple diagnostic and treatment applications including implant placement, sinus floor elevation, removal of (impacted) posterior teeth and/or root remnants, and orthognathic surgical procedures. Yet, it is challenging and time-consuming when manually performed on CBCT dataset. Recently, convolutional neural networks (CNNs) have proven to provide excellent performance in the field of 3D image analysis. Hence, this study developed and validated a novel automated CNN-based methodology for the segmentation of maxillary sinus using CBCT images. The proposed model provided a time-efficient, precise, and consistent automatic segmentation which could allow an accurate generation of 3D models for diagnosis and virtual treatment planning.
Researchers: Nermin Morgan, Adriaan Van Gerven, Andreas Smolders, Karla de Faria Vasconcelos, Holger Willems, Reinhilde Jacobs
Published on:
Project 11: Deep convolutional neural network-based automated segmentation of the maxillofacial complex from cone-beam computed tomography – A validation study
Automated segmentation of the maxillofacial complex could act as a potent alternative to the conventional segmentation techniques for improving the efficiency of the digital workflows. This approach could deliver an accurate and ready-to-print three dimensional (3D) models that are essential to patient-specific digital treatment planning for orthodontics, maxillofacial surgery, and implant placement.
Researchers: Flavia Preda, Nermin Morgan, Adriaan Van Gerven, Fernanda Nogueira-Reis, Andreas Smolders, Xiaotong Wang, Stefanos Nomidis, Eman Shaheen, Holger Willems, Reinhilde Jacobs
Published on:
Project 12: Deep convolutional neural network-based automated segmentation and classifcation of teeth with orthodontic brackets on cone-beam computed-tomographic images: a validation study
Tooth segmentation and classifcation from cone-beam computed tomography (CBCT) is a prerequisite for diagnosis and treatment planning in the majority of digital dental workfows. However, an accurate and effcient segmentation of teeth in the presence of metal artefacts still remains a challenge. Therefore, the following study aimed to validate an automated deep convolutional neural network (CNN)-based tool for the segmentation and classifcation of teeth with orthodontic brackets on CBCT images. The proposed CNN model outperformed other state-of-the-art algorithms in terms of accuracy and effciency. It could act as a viable alternative for automatic segmentation and classifcation of teeth with brackets.
Researchers: Khalid Ayidh Alqahtani, Reinhilde Jacobs, Andreas Smolders, Adriaan Van Gerven, Holger Willems, Sohaib Shujaat, Eman Shaheen
Published on:
Project 13: Three-dimensional maxillary virtual patient creation by convolutional neural network-based segmentation on cone-beam computed tomography images
This paper assessed the integrated segmentation of three convolutional neural network (CNN) models for the creation of a maxillary virtual patient (MVP) from cone-beam computed tomography (CBCT) images. We hypothesized that this integration would reveal a similar performance as the individual models, along with a robust interobserver agreement regarding time efficiency and consistency for creating a segmented MVP. Simultaneous automated segmentation of different structures could provide a valuable tool in clinical orthodontics, implant rehabilitation, and any oral or maxillofacial surgical procedures, in which visualizing MVP and its relationship with surrounding structures are necessary for an accurate diagnosis and patient-specific treatment planning.
Researchers: Fernanda Nogueira-Reis, Nermin Morgan, Stefanos Nomidis, Adriaan Van Gerven, Nicolly Oliveira-Santos, Reinhilde Jacobs, Cinthia Pereira Machado Tabchoury
Published on:
Project 14: Synergy between artificial intelligence and precision medicine for computer-assisted oral and maxillofacial surgical planning
The aim of this review was to investigate the application of artificial intelligence (AI) in maxillofacial computer-assisted surgical planning (CASP) workflows with the discussion of limitations and possible future directions. An in-depth search of the literature was undertaken to review articles concerned with the application of AI for segmentation, multimodal image registration, virtual surgical planning (VSP), and three-dimensional (3D) printing steps of the maxillofacial CASP workflows. The existing AI models were trained to address individual steps of CASP, and no single intelligent workflow was found encompassing all steps of the planning process. Segmentation of dentomaxillofacial tissue from computed tomography (CT)/cone-beam CT imaging was the most commonly explored area which could be applicable in a clinical setting. Nevertheless, a lack of generalizability was the main issue, as the majority of models were trained with the data derived from a single device and imaging protocol which might not offer similar performance when considering other devices. In relation to registration, VSP and 3D printing, the presence of inadequate heterogeneous data limits the automatization of these tasks. The synergy between AI and CASP workflows has the potential to improve the planning precision and efficacy. However, there is a need for future studies with big data before the emergent technology finds application in a real clinical setting.
Researchers: Sohaib Shujaat, Marryam Riaz, Reinhilde Jacobs
Published on:
Project 15: Convolutional neural network for automated classification of osteonecrosis and related mandibular trabecular patterns
The present study aimed to develop and validate a tool for the automated classification of normal, affected, and osteonecrosis mandibular trabecular bone patterns in panoramic radiographs using convolutional neural networks (CNNs). A dataset of 402 panoramic images from 376 patients was selected, comprising 112 control radiographs from healthy patients and 290 images from patients treated with antiresorptive drugs (ARD). The latter was subdivided in 70 radiographs showing thickening of the lamina dura, 128 with abnormal bone patterns, and 92 images of clinically diagnosed osteonecrosis of the jaw (ONJ). Four pre-trained CNNs were fined-tuned and customized to detect and classify the different bone patterns. The best performing network was selected to develop the classification tool. The output was arranged as a colour-coded risk index showing the category and their odds. Classification performance of the networks was assessed through evaluation metrics, receiver operating characteristic curves (ROC), and a confusion matrix. Furthermore, Gradient-weighted Class Activation Mapping (Grad-CAM) was employed to visualise class-discriminative regions. All networks correctly detected and classified the mandibular bone patterns with optimal performance metrics. InceptionResNetV2 showed the best results with an accuracy of 96 %, precision, recall and F1-score of 93 %, and a specificity of 98 %. Overall, most misclassifications occurred between normal and abnormal trabecular bone patterns. CNNs offer reliable potentials for automatic classification of abnormalities in the mandibular trabecular bone pattern in panoramic radiographs of antiresorptive treated patients.
Researchers: Soroush Baseri Saadi, Catalina Moreno Rabie, Tim van den Wyngaert, Reinhilde Jacobs
Published on:
Book chapters
Artificial Intelligence in Medicine (AIM)
Researchers: Maurício do Nascimento Gerhardt, Sohaib Shujaat, Reinhilde Jacobs
Published on:
Abstract:
The steep rise of digital dentistry and technological advancements have opened doors for the development of artificial intelligence (AI). For the past few years, AI-based applications in dentistry have been constantly evolving as highlighted by the increasing number of studies, and now it is slowly entering the clinical arena. As healthcare professionals, dentists need to diagnose, plan, and make clinical decisions in order to provide an adequate treatment and care for their patients. All these phases are time-consuming, observer-dependent, and subjected to human error. Currently, the studies applying AI in many dental specialties have validated its application for the purpose of diagnosis and clinical decision-making. Thus, the objective of AI is to combine the professional expertise with the computer-assisted systems to automatize complex tasks, mimic human cognitive skills, and retrieve information from digital data. Dental AI applications can be advantageous for all dental specialties including dentomaxillofacial radiology, restorative dentistry, oral and maxillofacial surgery, orthodontics, periodontics, prosthodontics, endodontics, and forensic dentistry.
This chapter provides an overview of the current state of the art of the AI applications in dentistry and its specialties.
RESEARCH COORDINATOR
Prof. dr. Reinhilde Jacobs
ADMINISTRATIVE COORDINATOR
Nele Vanloocke
© 2020 OMFS-IMPATH
