Reliability of Artificial Intelligence-Based Cone Beam Computed Tomography Integration with Digital Dental Images

Published: February 23, 2024

doi:

Ju-Heon Lee*¹, Nam-Ki Lee*¹, Bingshuang Zou, Jae Hyun Park⁴, Tae-Hyun Choi

¹Department of Orthodontics,Seoul National University Bundang Hospital, ²Division of Orthodontics, Faculty of Dentistry,University of British Columbia, ³Postgraduate Orthodontic Program, Arizona School of Dentistry & Oral Health,A. T. Still University, ⁴Graduate School of Dentistry,Kyung Hee University

Summary

A process of registering cone-beam computed tomography scans and digital dental images has been presented using artificial intelligence (AI) -assisted identification of landmarks and merging. A comparison with surface-based registration shows that AI-based digitization and integration are reliable and reproducible.

Abstract

This study aimed to introduce cone-beam computed tomography (CBCT) digitization and integration of digital dental images (DDI) based on artificial intelligence (AI)-based registration (ABR) and to evaluate the reliability and reproducibility using this method compared with those of surface-based registration (SBR). This retrospective study consisted of CBCT images and DDI of 17 patients who had undergone computer-aided bimaxillary orthognathic surgery. The digitization of CBCT images and their integration with DDI were repeated using an AI-based program. CBCT images and DDI were integrated using a point-to-point registration. In contrast, with the SBR method, the three landmarks were identified manually on the CBCT and DDI, which were integrated with the iterative closest points method.

After two repeated integrations of each method, the three-dimensional coordinate values of the first maxillary molars and central incisors and their differences were obtained. Intraclass coefficient (ICC) testing was performed to evaluate intra-observer reliability with each method's coordinates and compare their reliability between the ABR and SBR. The intra-observer reliability showed significant and almost perfect ICC in each method. There was no significance in the mean difference between the first and second registrations in each ABR and SBR and between both methods; however, their ranges were narrower with ABR than with the SBR method. This study shows that AI-based digitization and integration are reliable and reproducible.

Introduction

Three-dimensional (3D) digital technology has broadened the scope of diagnosis and planning for orthodontic or surgical-orthodontic treatment. A virtual head constructed from a facial cone-beam computed tomography (CBCT) image can be used to evaluate dentofacial and dental abnormalities, plan orthognathic surgery, fabricate dental wafers and implant surgical guides using computer-aided design and manufacturing¹^,²^,³^,⁴. However, CBCT scans have a low representation of dentition, including dental morphology and interocclusal relationship, which are due to their limited resolution and streak artifacts from dental restoration or orthodontic brackets⁵. Therefore, the dental features have been substituted on CBCT images with digital dental images (DDI), such as scanned casts or intraoral scan images.

For reliable integration of DDI on CBCT images, numerous studies reported various methods such as the use of fiducial markers⁶^,⁷, voxel-based⁸, and surface-based registrations (SBRs)⁹^,¹⁰. These procedures have their methods of using extraoral markers, multiple CBCT scans, and extra process steps such as cleaning metal artifacts on CBCT images. Regarding SBR accuracy, several previous studies reported errors ranging from 0.10 to 0.43mm⁹^,¹¹. In addition, Zou et al. evaluated intra-/inter-observer reliability and errors between a digital engineer and an orthodontist using SBR and reported the need for clinical experience and repeated learning¹⁰.

Artificial intelligence (AI) has been used to predict treatment outcomes¹² and digitize landmarks on cephalometric radiographs¹³ or CBCT images¹⁴^,¹⁵^,¹⁶, and some commercial software is currently available to assist in this process¹⁷. Accurate identification of anatomical landmarks on 3D images is challenging because of the ambiguity of flat surfaces or curved structures, areas of low density, and the wide variability of the anatomical structures.

AI-based, machine-learned automation can be applied not only for digitization but also for the integration of DDI and dentofacial CBCT. However, there is little research on the accuracy of an AI-based registration (ABR) compared to the existing surface-based method. To achieve more accurate outcomes of 3D skeletal and dental changes through bimaxillary orthognathic surgery, it is necessary to evaluate the accuracy of AI-based programs when merging CBCT and DDI. Therefore, this article presents a step-by-step protocol for digitizing and integrating CBCT and DDI with an AI-based registration (ABR) and to evaluate its reliability and reproducibility compared to that of SBR.

Protocol

This retrospective study was reviewed and approved by the Institutional Review Board of Seoul National University Bundang Hospital (B-2205-759-101) and complied with the principles of the Declaration of Helsinki. Digital Imaging and Communications in Medicine (DICOM) files from CBCT and DDI in Standard Tessellation Language (STL) format from the dental cast were utilized in the study. The need for informed consent was waived due to the retrospective nature of the study. 1. CBCT and Digit…

Representative Results

Here we described the integration process of CBCT and DDI using an AI-based program. To evaluate its reliability and reproducibility, a comparative study with surface-based registration (SBR) was conducted. It was determined that a minimum sample size of ten was required after a power analysis under correlation ρ H1 = 0.77, α = 0.05, and power (1−β) = 0.8018. A total of 17 sets of CBCT scans and digital dental images from orthognathic patients at Seoul National University Bund…

Discussion

Using the presented protocol, digitization of landmarks and integrating CBCT and DDI can be easily accomplished using machine-learned software. This protocol requires the following critical steps: i) reorientation of the head in the CBCT scan, ii) digitization of CBCT and DDI, and iii) merging CBCT images with the DDI. The digitization of five landmarks for the reorientation of the head is critical because it determines the 3D position of the head with reference planes in spatial areas. Three landmarks (R-/L-U6CP and R U…

Divulgazioni

The authors have nothing to disclose.

Acknowledgements

This study was supported by Seoul National University Bundang Hospital (SNUBH) Research Fund. (Grant no. 14-2019-0023).

Materials

G*Power	Heinrich Heine Universität, Dϋsseldorf, Germany	v. 3.1.9.7	A sample size calculuation software
Geomagic Qualify®	3D Systems, Morrisville, NC, USA	v 2013	3D metrology feature and automation software, which transform scan and probe data into 3D to be used in design, manufacturing and metrology applications
KODAK 9500	Carestream Health Inc., Rochester, NY, USA	5159538	Cone Beam Computed Tomograph (CBCT)
MD-ID0300	Medit Co, Seoul, South Korea Seoul, Korea	61010-1	Desktop model scanner
ON3D	3D ONS Inc., Seoul, Korea	v 1.3.0	Software for 3D CBCT evaluation; AI-based landmark identification, craniofacial and TMJ analysis, superimposition, and virtual orthognathic surgery
SPSS	IBM, Armonk, NY, USA	v 22.0	A statistic analysis program

Riferimenti

Plooij, J. M., et al. Digital three-dimensional image fusion processes for planning and evaluating orthodontics and orthognathic surgery. A systematic review. J Oral Maxillofac Surg. 40 (4), 341-352 (2011).
Badiali, G., et al. Virtual orthodontic surgical planning to improve the accuracy of the surgery-first approach: A prospective evaluation. J Oral Maxillofac Surg. 77 (10), 2104-2115 (2019).
Lee, C. Y., Ganz, S. D., Wong, N., Suzuki, J. B. Use of cone beam computed tomography and a laser intraoral scanner in virtual dental implant surgery: part 1. Implant Dent. 21 (4), 265-271 (2012).
Park, J. H., et al. Creation of digital virtual patient by integrating CBCT, intraoral scan, 3D facial scan: an approach to methodology for integration accuracy. J Craniofac Surg. 33 (4), e396-e398 (2022).
Almutairi, T., et al. Replacement of the distorted dentition of the cone-beam computed tomography scans for orthognathic surgery planning. J Oral Maxillofac Surg. 76 (7), 1561.e1-1561.e8 (2018).
Rangel, F. A., Maal, T. J., Berge, S. J., Kuijpers-Jagtman, A. M. Integration of digital dental casts in cone-beam computed tomography scans. ISRN Dent. 2012, 949086 (2012).
Yang, W. M., Ho, C. T., Lo, L. J. Automatic superimposition of palatal fiducial markers for accurate integration of digital dental model and cone beam computed tomography. J Oral Maxillofac Surg. 73 (8), 1616.e1-1616.e10 (2015).
Swennen, G. R., et al. A cone-beam computed tomography triple scan procedure to obtain a three-dimensional augmented virtual skull model appropriate for orthognathic surgery planning. J Craniofac Surg. 20 (2), 297-307 (2009).
Noh, H., Nabha, W., Cho, J. H., Hwang, H. S. Registration accuracy in the integration of laser-scanned dental images into maxillofacial cone-beam computed tomography images. Am J Orthod Dentofacial Orthop. 140 (4), 585-591 (2011).
Zou, B., et al. Accuracy of a surface-based fusion method when integrating digital models and the cone beam computed tomography scans with metal artifacts. Sci Rep. 12 (1), 8034 (2022).
Lin, H. H., et al. Artifact-resistant superimposition of digital dental models and cone-beam computed tomography images. J Oral Maxillofac Surg. 71 (11), 1933-1947 (2013).
Park, J. H., et al. Use of artificial intelligence to predict outcomes of nonextraction treatment of Class II malocclusions. Semin Orthodontics. 27 (2), 87-95 (2021).
Bao, H., et al. Evaluating the accuracy of automated cephalometric analysis based on artificial intelligence. BMC Oral Health. 23 (1), 1-10 (2023).
Gillot, M., et al. Automatic landmark identification in cone-beam computed tomography. Orthod Craniofac Res. , (2023).
Blum, F. M. S., et al. Evaluation of an artificial intelligence-based algorithm for automated localization of craniofacial landmarks. Clin Oral Invest. 27, 2255-2265 (2023).
Al-Ubaydi, A. S., Al-Groosh, D. The validity and reliability of automatic tooth segmentation generated using artificial intelligence. The Scientific World Journal. 2023, 5933003 (2023).
Urban, R., et al. AI-assisted CBCT data management in modern dental practice: Benefits, limitations and innovations. Electronics. 12 (7), 1710 (2023).
Faul, F., Erdfelder, E., Buchner, A., Lang, A. G. Statistical power analyses using G*Power 3.1: tests for correlation and regression analyses. Behav Res Methods. 41 (4), 1149-1160 (2009).
Landis, J. R., Koch, G. G. The measurement of observer agreement for categorical data. Biometrics. 33, 159-174 (1977).
Uechi, J., et al. A novel method for the 3-dimensional simulation of orthognathic surgery by using a multimodal image-fusion technique. Am J Orthod Dentofacial Orthop. 130 (6), 786-798 (2006).
Nkenke, E., et al. Fusion of computed tomography data and optical 3D images of the dentition for streak artifact correction in the simulation of orthognathic surgery. Dentomaxillofac Radiol. 33 (4), 226-232 (2004).
Lagravère, M. O., et al. Intraexaminer and interexaminer reliabilities of landmark identification on digitized lateral cephalograms and formatted 3-dimensional cone-beam computerized tomography images. Am J Orthod Dentofacial Orthop. 137 (5), 598-604 (2010).

Play Video

PDF

DOI

DOWNLOAD MATERIALS LIST

Citazione di questo articolo

Lee, J., Lee, N., Zou, B., Park, J. H., Choi, T. Reliability of Artificial Intelligence-Based Cone Beam Computed Tomography Integration with Digital Dental Images. J. Vis. Exp. (204), e66014, doi:10.3791/66014 (2024).