Background
The development of digital scanners brings about a lot of changes to the dental treatment environment. As a non-invasive method without radiation exposure, the digital scanning method is expected to reduce treatment time and improve the quality of treatment by allowing frequent evaluations according to needs [
1]. It is even possible to know the state of occlusal contact by utilizing digital imagery by scanning models using an intraoral scanner. The use of the intraoral scanner offers the advantage of excluding the necessity of the impression process, production of the master cast, errors in the process of mounting on the articulator and errors due to the thickness of the occlusal evaluating paper. The manners of recording the interocclusal relationship include scanning the impression material of the maxillo-mandibular occlusal relationship, and scanning the buccal portion at centric occlusion. There is a risk of movement of the impression material during the scanning process in the former method, so the latter method which offers useful information in the analysis of the state of occlusion in an actual patient without the need of any impression materials by directly capturing the occlusion and buccal contact points is preferred.
Digital impressions are divided into a direct method and an indirect method. The direct method is where digital imagery is obtained by scanning the oral environment of the patient, and the indirect method is where images in the form of digital models are obtained by scanning plaster models. Most researches evaluate the reproducibility and accuracy of the scanner by comparing the virtual models obtained from the indirect method with the plaster models. The size of tooth, width of arch etc. are usually compared by comparing the measured values of the plaster model with use of a digital caliper to the values obtained from scanned images [
2‐
4]. But there are many factors which may cause errors in these methods, so as of recently there has been an introduction of analytical methods which employ 3-dimensional superimposition for comparison and analysis [
5]. Images obtained from scanners are not only used in the general production of dental prostheses, but also as guides in implant operations and the production of implant prostheses, and more recently, in the diagnosis and planning of orthognathic surgery and orthodontic setups [
6‐
10]. This signifies that operations that analyze the manner of occlusal contacts and the state of the occlusal plane of the complete maxillary and mandibular dentition using digital images has become commonplace.
As a result, new methods that analyze the occlusal points, the occlusal area and the occlusion have surfaced as a research topic. The T-scan and the Prescale are representative [
11]. With the T-scan, the patient is instructed to carry out a series of occlusal movements such as lateral movements and anterior movements with a sensor placed in between their upper and lower arches, which is then recorded making it possible to analyze the order of occlusion according to the duration and strength of pressure applied on the tooth. The T-scan is recognized as a reproducible method in the analysis and evaluation of occlusal contacts at maximum intercuspation [
12‐
15]. The Prescale system which was developed in Japan in that early 1990s consists of a pressure sensitive film in which a pressure between 5Mpa and 150Mpa applied to it will result in the destruction of micro capsules within the film, and a red color appears as a colorless dye mixes with a developing solution in the film. The color becomes darker as the pressure applied to it is higher, and the difference in the density of the color is analyzed. Hattori et al. [
16] stated that the Prescale was meaningful as scientific data since it is possible to compare data obtained in a large scale research such as cohort studies. Suzuki et al. [
17] stated that the speed and duration of the strength of occlusion applied to the Prescale sheet does not influence any changes in its color, and that the Prescale sheet does not get influenced by the moisture in the oral cavity. Also, according to the research conducted by Hidaka et al., [
18] it was reported that there was a strong linear relationship between the load applied to the Type-R 50H sheet and the load deciphered from the sheet, making the Prescale it a trustworthy system in measuring the occlusal force.
Orthodontic treatment is the solution of malocclusion and the creation of functional occlusion. For 3-dimensional imagery to be generalized in orthodontic treatment, evaluation of the accuracy of the virtual occlusion obtained from the intraoral scanner is required. Research on the occlusal contact of restorations obtained from single tooth or partial arch scans are ongoing, but research on the occlusal contact of the full arch using virtual occlusion is meager and its accuracy has not been proven [
19‐
21]. Before this study was conducted, clinically when the virtual occlusion was constructed, errors were commonly seen in certain area. Even though there was a lack of contact in the anterior portion of the maxilla and mandible, contact between the two arches was indicated. But because of the limited space when directly scanning the patient’s mouth leads to a decrease in accuracy, [
22‐
25] a cast study which can control these conditions and increase reproducibility is in need. Therefore, the aim of this study was to evaluate the accuracy of virtual occlusions obtained by intraoral scanners using plaster dental models in vitro, by comparing virtual occlusal contact area with actual occlusal contact area obtained by prescale method.
Discussion
Currently, intraoral scanners are not only used in obtaining images of the single tooth and the dental arch, they are also utilized in the diagnosis and planning of orthognathic surgery and orthodontic setups [
7,
9,
10,
29]. Using digital images obtained from scans to analyze the state of occlusal contact as well as the morphology of the occlusal surface of the complete maxillo-mandibular dentition has become more common. For intraoral digital scanners to be used more generally, it must be possible to obtain precise scan images of the full arch and dentition, and the virtual occlusion that results from those images must be able to reproduce the actual occlusion.
This study measured and compared the occlusal contact area of actual models obtained from the Prescale and the virtual occlusion obtained from scanned images. The results show that in the molar portion, there was little difference in the occlusal contact area obtained from the virtual occlusion and that obtained from the Prescale and there was no statistically significant difference. (p > 0.05) In the molar portion, the ratio of the occlusal contact area of the virtual occlusion to the Prescale showed a high degree of correlation, almost forming a linear line.(R = 0.964, B = 0.917) There was a statistically significant difference in the occlusal contact area of the virtual occlusion and the Prescale in the anterior portion, (p < 0.05) and compared to the occlusal contact area obtained from the Prescale, the occlusal contact area obtained from the virtual occlusion was overestimated. During the course of the research, there were a lot of cases where occlusal contact was expressed in the anterior portion of the virtual occlusion even though there was no contact in the same region according to the Prescale.
Compared to a single tooth or a portion of the arch, scanning the full arch is technically difficult and there is more chance of error when obtaining the images [
30]. It has been mentioned in a number of previous studies that the magnitude of these errors are clinically acceptable [
5,
30,
31].
The results of superimposition carried out by most of the previous studies in the literature aimed at evaluating the accuracy of the intraoral scanner show a lower degree of accuracy in the distal portion of the molars and the incisal surface of the anterior teeth [
5,
31,
32]. These errors occur due to the presence of sharp curves in the distal portions of the molars and the incisal portions of the anterior teeth which cause diffraction of light resulting in errors when obtaining images [
27,
33]. In other words, precise scan images of complex structures with the presence of undercuts are difficult to obtain. Even in this study, precise scan images could not have been obtained in the anterior portion due to the diverse morphology and tooth crowding present during the individual scanning of the maxillary and mandibular models.
The algorithm involved in the manner of registration of 3-dimensional images of the intraoral scanner could have been the cause of the results in the anterior portion. The scanner used in this study employs the best fit algorithm. There have been many studies on the best fit algorithm [
34,
35]. The most representative problem with this method is that the errors that occur in the beginning continue to add up as the scanning process continues. Solaberrieta et al. [
36] recommended that in order to reduce the errors in the best fit algorithm, when employing the buccal scan method to produce the virtual occlusion, instead of scanning the full arch, to only obtain three precise scan images (images of the bilateral molar area and the frontal anterior image) of a width of 24 mm and a height of 5 mm. They stated that if these three images can be obtained, a virtual model occlusion that is closest to the actual model occlusion was possible. The important point was that the distance among the sections should be as large as possible, and that precise images of the left and right molars were imperative. In reality, with the models in maximum intercuspation, when the buccal surface is scanned from the molar to the anterior portion, even before the full arch is scanned, the program completes the occlusion. Due to the best fit algorithm, as scanning progressed from the molar portion to the anterior portion, it seems that the errors added up as the scanning process continued to present itself with overestimation in the anterior portion in this study.
Intraoral scanner used in this study employs ultrafast optical sectioning technology based on confocal microscopy, taking more than 3000 2D images per second and then combining them into 3D. In the process of creating the virtual occlusion, the scanner determines the relative position of the maxilla and the mandible through the process of scanning, and depends on the images of the teeth that are already registered, as well as the images of the surface of the maxillary and mandibular teeth in intercuspation. But these images do not include any information on the direction, position or angle in a 3-dimensional space [
21,
34]. Due to the lack of 3-dimensional information, it is difficult to position the scanned images of the maxilla or the mandible when there is a partial or complete edentulous area. And detailed registration of superimposed images of two scans is also difficult [
34]. As it is shown in the results of this study, when the virtual occlusion is formed from the registration of the two images, the anterior portion is shown to contact more closely than in reality. Also, the maxillo-mandibular labial images that are superimposed as a result of the diverse overbite and overjet in the anterior labial portion may not have supplied adequate information in the registration of the maxilla and mandible, which could be the cause of another form of error. In reality, to solve these problems, methods for precise registration are being suggested and effort in developing such software is underway [
37,
38]. Research uses the virtual articulator to construct the virtual occlusion from scanned images [
37,
39]. Delong et al. [
37] demonstrated the possibility of reproducing the functions of the virtual articulator and actually obtaining acceptable results by comparing its accuracy under various conditions, and creating the necessary software to analyze the occlusion from virtual models.
The Prescale which was used as a control group in this study makes it possible to analyze the occlusal contact area of the full arch quantitatively, and is known as a trustworthy system in measuring the occlusive force. It can be with better use as it enables to compare scientific data when large data is collected [
15,
17]. The Prescale is a good means to quantitatively evaluate the occlusive force, but it has its limitations in the form of the deformation of its sensors and its thickness. The thickness of the film may interrupt movement onto intercuspation and it has been reported that there is a limitation as a result of the distortion of the pressure sensor [
15].
But currently, of the methods of measuring the occlusive force, since there is no golden standard that can represent the occlusal contact points, the Prescale which is a scientific method that can display the occlusive force was used in this study.
Being conducted with plaster models, the results of this study could be different from clinical tests. Errors of the Prescale resulting from the non-uniform pressure upon closing or the position of the head, by applying uniform pressure to the models with the Instron were avoided but it was not possible to express the maximum intercuspation through the opening and closing of the mandible in an actual oral cavity. However it can be a good milestone to develop the software for 3d intraoral scanner. And it can examine the difference between virtual and real occlusal by scanning system as it reduces the errors from intraoral scanned images and imaging skills.
Conclusion
This study evaluated the accuracy of the jaw relation record through the buccal scan method in a state of occlusion by comparison of the occlusal contact area obtained by the Prescale and a 3D intraoral scanner, and whether it could be used clinically. The tooth groups were divided into a molar portion, a premolar portion, and an anterior portion. The occlusal contact area obtained from the Prescale and the 3D intraoral scanner was measured, ultimately comparing the virtual occlusion and the actual occlusion.
In the molar portion and the premolar portion, there was no statistically significant difference between the occlusal contact area obtained from the Presacle method and the scanning method (p > 0.05). There was a statistically significant difference between the two methods in the anterior portion where overestimation was observed. (p < 0.05).
There was no statistically significant difference between the actual occlusion and the virtual occlusion using 3D scan images in the molar portion, but there was statistically significant difference between the two methods in the anterior portion, which presented with overestimation of the occlusal contact area obtained from the scanning method. The virtual occlusion using scanned images can be used as a reference in the diagnosis and modification of the occlusal contact. But there are limitations to the scanning method, therefore this should be taken into consideration when applying it in the clinical environment. Software and scanning techniques that can more precisely reproduce the patient’s jaw relation must be developed.