Temporary Skeletal Anchorage Devices and Cone Beam Tomography in Orthodontics—Current Application and New Directions of Development

Abstract

Continuous progress in dentistry and orthodontics is crucial to ensuring high-quality diagnosis and treatment of patients, especially since malocclusions occur in up to half of the population. In addition to limiting the physiological functions of the masticatory system, they are often an aesthetic defect that may directly affect the well-being and even self-esteem of patients, especially in their teenage years and early adulthood. A holistic model of perceiving and treating orthodontic diseases, such as the Biocreative Orthodontic Strategy, focusing not only on the correction of the defect itself but also taking into account the least possible interference in the physiology of the masticatory system, limiting the use of appliances to a minimum, and taking into account the patient’s preferences, is a special alternative to conventional therapeutic models. In this review, we are presenting the current knowledge regarding the applications of temporary skeletal anchorage devices (TSAD) and cone beam computed tomography (CBCT) in orthodontics.

1. Introduction

Orthodontics is a dynamically developing field of dentistry, used not only to improve the aesthetics of the bite but also to correct serious defects of the masticatory system that disturb its functions [1,2,3]. According to the American Association of Orthodontists (AAO), 50% of people have malocclusions that qualify for orthodontic correction, of which 10% are cases that require medical intervention [4,5,6]. According to the WHO, malocclusion is one of the three most important dental diseases, right after periodontitis and caries. Estimates of its prevalence range from 39% to 93% in children and adolescents [7,8]. Malocclusions were classified by E. Angle and divided into four classes (normal bite, class I, class II, and class III) based on the relationship between the position of the upper and lower first molars. Malocclusion is not only a cosmetic defect but also, according to the position of the World Dental Federation (FDI), may affect oral health and contribute to an increase in the incidence of tooth decay, periodontitis, risk of injuries, and difficulties in chewing, swallowing, breathing, and speaking, which emphasizes the importance of developing orthodontics as an integral part of dentistry [9,10]. One type of orthodontic appliance widely used in orthodontics are TSADs—temporary skeletal anchorage devices. The main advantage of this type of device is the possibility of mounting to the palatal bone or alveolar ridges, most often using mini-plates and mini-screws, which ensures stable anchoring without the need to use a rest for the premolars and molars, which in turn could lead to their undesirable displacement. TSAD is often used in the en masse retraction of anterior teeth after premolar extraction in the treatment of bimaxillary protrusion. Thanks to this method, controlled and relatively quick alignment of the front teeth is possible [11,12,13,14,15,16]. One of the developed treatment strategies using TSAD is BOS—Biocreative Orthodontic Strategy, the main goal of which is to correct malocclusions with the least possible interference in the physiological functions of the masticatory system. It assumes personalized adjustment of therapy and appropriate orthodontic appliances to the patient’s age, anatomical differences, and personal preferences. It is important that during treatment, among others, bimaxillary protrusion does not lead to a change in the location of the posterior teeth; therefore, partially osseointegrable C mini-implants are preferred as anchoring devices. BOS is a holistic strategy for the treatment of malocclusions with great potential and effectiveness [17,18,19,20,21,22]. Imaging diagnostics plays an important role in the modern diagnosis of malocclusion. In recent years, diagnostic methods that allow 3D imaging of structures have become increasingly important. Such techniques include the following: computed tomography (CT), cone beam computed tomography (CBCT), computer-aided design/computer-aided manufacturing (CAD/CAM), micro computed tomography (MCT), 3D laser scanning, structured light technology, stereophotogrammetry or systems 3D surface imaging (3dMD), 3D facial morphometry (3DFM), adjustable aperture computed tomography (TACT), and magnetic resonance imaging (MRI) [23,24,25,26]. In dentistry and orthodontics, CBCT cone beam computed tomography is preferred because it allows precise imaging of a given part of the facial skull with several times less X-ray radiation than in the case of classic CT, which is important especially in the pediatric population. Other advantages of CBCT include the short duration of the examination and relatively low cost [27,28,29,30,31,32,33]. The main difficulty associated with performing the examination in children is the frequent acquisition of images distorted by artifacts resulting from head movement. To minimize the occurrence of this problem, it is considered to perform the examination in a position as comfortable as possible for the child with gentle stabilization of the head. CBCT in children helps in imaging such pathologies as an ectopic maxillary canine, retained second premolar, cleft palate, or dental root resorption, and most importantly, in preparation for orthodontic treatment. CBCT is also used in the imaging of malocclusion and preparation for TSAD therapy. The bone and soft tissue thickness measurements obtained facilitate the selection of an appropriate site for anchoring the appliance, and the small amounts of radiation during the examination allow frequent follow-up imaging during treatment [2]. Due to the wide use of cone tomography in dentistry, new technologies are constantly being developed to increase the effectiveness and precision of imaging. In recent years, a new direction has been the use of artificial intelligence, including convolutional neural networks and deep learning models [34,35,36,37]. The use of AI significantly shortens the work of clinicians by automatically analyzing and describing structures on CBCT images. It can help prepare for malocclusion correction strategies by marking specific anatomical points or for implant placement or the attachment of temporary anchorage devices by determining alveolar bone thickness. The most important advantage of using AI in the analysis of CBCT results seems to be saving orthodontists’ work and time, as well as the ability to compare data from many images at once [38,39,40,41,42]. This technology is new and under intensive development, so it will likely be used in other aspects of diagnostic imaging in the future. Further research is certainly necessary to fully exploit the potential of AI in this area. This article is a review of information available in the PubMed and Web of Science databases on the current use of temporary skeletal anchorage devices (TSAD and CBCT) computed cone tomography in orthodontics, including new directions of development in their field (

Table 1. Illustrating the number of articles available in the PubMed and Web of Science databases after entering the phrases “CBCT AND orthodontics”, “CBCT AND artificial intelligence”, “TSAD AND orthodontics”, “TSAD AND en masse retraction”, and “Biocreative orthodontic strategy”. The largest number of results was achieved when searching for the phrase “CBCT AND orthodontics” in the PubMed search engine (2701 results), of which almost half of the articles were published in the last 4 years (2020–2024). A smaller but still significant number was achieved by searching for “CBCT AND artificial intelligence”—more results in Web of Science (266), the vast majority of which concern the years 2020–2024 (251 results). In other cases, the number of articles received is several to a dozen at most, as in the case of “TSAD AND orthodontics”, where most of the articles were written in the last 4 years (8/13 and 5/7). It is worth paying attention to the rapidly growing interest in the use of artificial intelligence in CBCT cone beam tomography in recent years. Access on 2 March 2024.

Keywords	Resuts in PubMed		Results in Web of Science
	Total	Time 2020–2024	Total	Time 2020–2024
“CBCT AND orthodontics”	2701	1342	1101	512
“CBCT AND artificial intelligence”	234	215	266	251
“TSAD AND orthodontics”	13	8	7	5
“TSAD AND en masse retraction”	2	1	3	2
“Biocreative orthodontic strategy”	10	3	2	2

). This article emphasizes the importance of technological advancements, such as TSADs and CBCT, in improving orthodontic care and patient outcomes.

2. The Use of TSADs in Orthodontics

TSADs (temporary skeletal anchorage devices) are widely used in orthodontics to correct malocclusions. They support the retraction of anterior teeth, protraction of posterior teeth, as well as vertical and lateral control of the jaw teeth [15]. They are used for the treatment of mild and moderate dent alveolar dysplasia, as well as for transverse correction with minimal widening of the maxillary alveolar process segments, for closing the bite with intrusion of molars, for correcting a cross bite in combination with a small segmental osteotomy, and for correcting anterior tilts [17]. TSADs support the correction of malocclusions using segmented archeries that affect the front and back teeth separately, thus maintaining greater control of their movement. The advantage of TSAD is its strong and durable anchorage, the loss of which has so far been a common problem when segmented archeries are used [25]. One of the main applications of TSAD is the retraction of anterior teeth after premolar extraction in the treatment of bimaxillary protrusion [11,12,13,14]. Protrusions are as follows: bimaxillary or maxillary-alveolar protrusions are among the most common malocclusions treated orthodontically all over the world [11,12]. One- or two-maxillary extraction of the first or second premolars is often necessary to create space for the retraction of the incisors and canines [11,14]. Sometimes removing four teeth is the most effective solution, although this solution is not always accepted by patients [14,43]. There are two main methods of retraction of anterior teeth, i.e., the two-step method—“two-step anterior teeth retraction” (TSR), where first the canines are retracted and then the incisors, which is usually combined with conventional anchoring of the braces, and the one-stage method—“en masse retraction (ER), where the canines and incisors are simultaneously retracted. Many articles focus on the use of TSAD in this method, comparing the therapeutic and aesthetic effects achieved, the duration of treatment, and its impact on the skeleton, teeth, and soft tissues [11,12,13,14,15,43,44,45,46]. In the case of the two-stage method, the advantage is that fewer teeth are incorporated into the anchor unit than in the case of ER, so it is more difficult to destabilize it, and standard anchoring methods can usually be used. However, the separate introduction of incisors and canines into the arch has its limitations, i.e., there is a risk that the canines become excessively rotated and tilted, and the therapy itself is prolonged [11,44,45]. In turn, the use of en masse retraction using TSAD and T-loop wires allows for much more precise control over tooth movements and ensures optimal distribution of forces acting on them, thanks to which it is possible to supervise the therapeutic process with greater accuracy and predict its effect. However, this method often requires reinforced anchorage due to the larger number of teeth incorporated into the fixation unit [44,46]. The use of TSADs allows for tooth movement control regardless of attachment to other teeth [13]. There are usually four types of fixations of TSADs to the bone, i.e., miniscrews, miniplates, implants, and onplants, of which miniscrews and miniplates are most often used [15]. The thickness of the palatal bone or alveolar ridge plays an important role in the stability of the anchor device—the screw length should be carefully adjusted to avoid bone perforation (although according to some sources, such a situation is desirable due to the supposed more permanent fixation [16]). This thickness must be at least 1 mm. Appropriate bone density and mucosa thickness are also important for maintaining the plate [15]. In the case of en masse retraction, miniscrews and miniplates are usually used, attached to the palatine bones. This method allows for maximum stable anchorage, which is crucial for the success of therapy in the case of one-stage retraction [15]. It is also important that when the front teeth are retracted, the position of the rear teeth changes only minimally. The camera can be attached indirectly or directly. The indirect method also uses arches, lingual or transpalatal, attached to premolars or molars [47,48]. Cone-beam computed tomography (CBCT) imaging is helpful in selecting a suitable site for anchoring the device, taking into account the appropriate bone thickness. This method, described in more detail later in this article, allows precise evaluation of the craniofacial region, including soft tissues, and treatment planning with TSAD [2]. The effectiveness of TSADs as a device maintaining a sufficiently stable anchorage of the appliance using miniscrews was determined to be over 90%, which is a much higher result compared to devices mounted in the buccal interdental space [49,50]. In the work of Kim K. et al. [51], they assessed differences in the thickness of the buccal alveolar bone in six zones in patients with different morphological structures of the facial skeleton in order to determine a safe location for TSAD installation. It has been determined that the safest, universal space for insertion of the fixation miniscrew is 11 mm from the alveolar crest between the maxillary first and second molars and at the medial root of the second molar. Moreover, it is stated that the placement of the miniscrew in the palatal suture, age (especially > 15 years), and skill of the surgeon predispose to the success of the therapy [15,49]. In a study by Kim S.H. et al. [52], the effects of retraction of six anterior teeth using TSADs as the sole source of anchorage after extraction of premolars were assessed in 17 patients. It was found that the goal was achieved without the need for lateral appliances, but in some cases a loss of anchorage comparable to traditional methods was observed. In turn, the study by Antoszewka-Smith J. et al. [53] compared the effectiveness of TSAD and conventional methods of anchoring the appliance when closing the arch by retraction of the anterior teeth. The obtained results indicated better protection of the device anchorage compared to traditional methods, as well as greater incisor retention using TSADs. Only the degree of tilting of both molars and incisors during therapy did not differ significantly between the two methods. The authors also state that the presented results should be treated with some caution and that more retrospective studies are necessary for a complete analysis. Undoubtedly, however, the use of temporary skeletal anchorage devices has great potential for effectiveness in various orthodontic conditions. This method is still developing, and many research teams are working on its new applications (Figure 1, Figure 2 and Figure 3).

3. TSAD and the Principles of the Biocreative Orthodontic Strategy (BOS)

Any form of orthodontic treatment involves more or less interference with the natural development and function of the masticatory apparatus. Before deciding on a particular therapy, it is important to analyze its gains and losses. Most orthodontic interventions are applied at developmental ages, when the skeleton is most malleable. For this reason, it is so important to limit the adverse effects of treatment, which can permanently affect the patient’s future life, and to minimize radical forms of therapy. This is why the Biocreative Orthodontic Strategy (BOS) was developed between 1979 and 1998 [17]. BOS is a strategy for the treatment of various orthodontic defects that is based on the least possible interference in the physiological functions of the masticatory apparatus and avoiding unfavorable bite changes during therapy (Figure 4). It assumes the use of TSAD as a tool to help develop and maintain an appropriate occlusion pattern. The emphasis is on the physiological properties of the skeleton depending on the moment of development, anatomical variability of the patient, maintaining maximum patency of the upper respiratory tract, and appropriately simplified biomechanics of forces acting on the teeth, related to minimizing the number of orthodontic appliances worn [17]. It is also important not to attach fixed appliances to the posterior teeth during the retraction of the front teeth of the maxilla, which could lead to undesirable changes in occlusion [22]. Particularly in the pediatric group, it is important to ensure that the jaws, nasal bones, and mandible are properly aligned with each other to promote proper airway formation. For this purpose, a horseshoe-type maxillary expander and tongue elevator are used, which, through Merrifield’s directional forces, allow control of mandibular rotation with proper tongue positioning for maximum patency [17]. According to BOS, the treatment effect includes the following: biomechanics of forces acting on the teeth, patient/parent cooperation, appropriate limits of the alveolar ridge, condylar position, target movement of bones and teeth, as well as minimizing the amount of TSAD used, observation of the tongue position, and the effectiveness of RME (rapid maxillary expansion) [17,18].

4. TSAD: The Use of C Minimplants and NiTi Arches

In the BOS strategy, miniscrews with a diameter of 1.2–2.0 mm and a length of 6–12 mm are most often used for TSAD fixation for distalization, intrusion, protraction, or indirect anchoring, as well as partially bone-integrable mini-implants (C-implant) and miniplates with C tube [17,19]. The latter allows for the multidirectional application of high-intensity force, consistent with the assumptions of biomechanics, opposing torque [20,21,54]. At the same time, wire arches can be used, such as conventional C-arches, which, in combination with a C-implant, make it possible to minimize friction and load on the posterior teeth during the retraction of the incisors and canines [22]. Implant C, thanks to its partial integration into the alveolar bone, provides strong support for the entire structure. The wire arch is attached to the C implant holes, and the apical arches are attached to it. In this way, an intrusion force is exerted on the anterior teeth, preventing their lingual tilt during en masse retraction [55]. In the study by Chung K.R. et al. [56], nickel–titanium arches with reverse Spee curvature were used, attached to the C implant, which allowed the device to remain stable during therapy and avoid changes in the position of the posterior teeth. The on-lay nickel–titanium arch with reverse curve Spee exerts torque on the C-implant without loosening the screws and creates forces responsible for rotation and control of the vertical position of the incisors. Moreover, the size of the wire archwire can be easily adjusted and changed [54]. In a study by Mo S.S. et al. [54], a three-dimensional model of maxillary teeth, periodontal ligaments, and alveolar processes was constructed after extraction of first premolars. A mini-implant was then placed with a 0.8 mm hole between the second premolar and first molar and an inversely curved NiTi archwire from the head of the C implant to the point between the central incisors and to the segmental arch of the six anterior teeth, using the mini-implant as a posterior orthodontic tube. The influence of different sizes of retraction hooks placed between the lateral incisor and the canine on the torque control and the effect of forces of varying intensity on the NiTi overlay arch were examined. No braces or brackets were placed on the back teeth. The results of the study confirmed that the forces caused by variable-sized retraction hooks on the NiTi arch caused measurable, desired intrusion of the incisors and extrusion of the canines, both tooth crowns and roots. It has been proven that the desired effect of en masse retraction of anterior teeth can be achieved by using a partially osseointegrable C mini-implant as the only source of anchorage, a NiTi overlay arch, and retraction hooks of a specific size without interfering with the position of the posterior teeth of the jaws (biocreative therapy type II). [54]. Before modern nickel–titanium archwires were used, conventional stainless steel archwires (SS archwires) were used to retain anterior teeth. However, they had certain limitations: their skillful preparation and installation required additional training and training from orthodontists, and during treatment, due to the high coefficient of deflection under load of the SS wires, they required frequent inspections and time-consuming adjustments [22,57,58,59]. Therefore, Mo S.S. et al., in their work [22], decided to create a new hybrid Biocreative retractor (CH-retractor), which consists of a nickel–-titanium (NiTi) section covering the front teeth, from canine to canine, and rear sections of the SS arch, whose task is to transfer the force about the appropriate vector for the front teeth. The only source of anchoring the device is partially bone-integrable C implants and C-tube miniplates, which do not require the involvement of the posterior teeth. The CH-retractor can be used in the treatment of maxillary-alveolar protrusion, as well as other malocclusions, and also in the case of very large crowding of front teeth, where appropriate NiTi arch attachments are used; they are attached to the central incisors and canines [22,60]. Distal forces act on the canines to make room for the lateral incisors, which are also attached to the archwire with a steel ligature to ultimately adapt to its physiological shape. This modern method allows you to start effective retraction of front teeth along with alignment and closing of the arch from the first day of therapy. Moreover, in the same work [22], scientists assessed the retraction model of anterior teeth using a hybrid retractor using the finite element method (FEM). The obtained results confirm the effectiveness of the device, which made it possible to achieve the desired position of six front teeth using nickel–titanium front archwires and SS archwires with specific peak bending values and without the need to install additional structures on the rear pre- and molars. The cited studies confirm the effectiveness of the use of modified TSADs in the form of mini-implants or mini-C plates as the only source of anchorage in the correction of malocclusions using retractors without having an adverse effect on the position of the posterior teeth. List of mentioned authors and works related to TSAD and BOS (

Table 2. List of mentioned authors and works related to TSAD and BOS.

Reference Number in Our Work	Authors	Article Title	Publication Year	Type of Work
[22]	Mo S.S. et al.	“Finite element study of controlling factors of anterior intrusion and torque during Temporary Skeletal Anchorage Device (TSAD) dependent en masse retraction without posterior appliances: Biocreative hybrid retractor (CH-retractor)”	2020	Finite element study
[51]	Kim K. et al.	“Enhanced artificial intelligence-based diagnosis using CBCT with internal denoising: Clinical validation for discrimination of fungal ball, sinusitis, and normal cases in the maxillary sinus”	2023	Clinical validation
[52]	KIm S.H. et al.	“Analysis of temporary skeletal anchorage devices used for en-masse retraction: a preliminary study”	2009	Preliminary study
[53]	Antoszewska-Smith J. et al.	“Effectiveness of orthodontic miniscrew implants in anchorage reinforcement during en-masse retraction: A systematic review and meta-analysis”	2017	A systematic review and meta-analysis
[54]	Mo S.S. et al.	“Factors controlling anterior torque with C-implants depend on en-masse retraction without posterior appliances: biocreative therapy type II technique.”	2011	Finite element study
[56]	Chung K.R. et al.	“Anterior torque control using partial-osseointegrated mini-implants: biocreative therapy type II technique”	2008	Case report

).

5. Cone Beam Computed Tomography—Current Use in Orthodontics

Three-dimensional imaging is currently a key element in the diagnosis and treatment of malocclusions. The methods used in 3D technology include the following: computed tomography (CT), cone beam computed tomography (CBCT), computer-aided design/manufacturing (CAD/CAM), micro computed tomography (MCT), 3D laser scanning, structured light technique, stereophotogrammetry or 3D surface imaging systems (3dMD), 3D facial morphometry (3DFM), adjustable aperture computed tomography (TACT), and magnetic resonance imaging (MRI) [23,24,25,61,62]. Three-dimensional imaging allows you to personalize therapy through accurate measurements of the facial skull and the creation of perfectly fitting overlays, braces, guides, and orthodontic splints [24]. It becomes possible to accurately visualize the structure of teeth and their position relative to each other [63,64,65]. CBCT, i.e., cone tomography, was introduced to dentistry and orthodontics at the turn of the 20th and 21st centuries, being a breakthrough three-dimensional imaging method, offering a significantly wider range of diagnostic possibilities than 2D methods [66,67]. With a lower radiation dose than classic CT, it allows for a precise assessment of the structure of soft tissues [2,68]. It is used in the imaging and treatment of such defects and diseases as impacted teeth, supernumerary teeth, resorption of tooth roots, cleft upper lip and palate, pathologies in the jaws and mandible, temporomandibular joint, and alveolar processes. Diagnosing respiratory disorders such as sleep apnea is also important. CBCT is used to perform accurate measurements of the facial skeleton, necessary to assess malocclusions, as well as to monitor the course of orthodontic treatment, also with the use of TSAD [69,70,71,72,73]. It is also useful in determining the thickness of the alveolar bone and gums, as well as in creating spatial models that help in selecting a place for fixing the TSASD [74]. CBCT works on the principles of ordinary computed tomography, where the patient is exposed to an X-ray tube on one side, and on the other, there is a detector that measures the amount of radiation absorbed by the tissues. The radiation beam falling on the tissues has the shape of a cone, and the examination allows for volumetric imaging of a given structure [29]. Both the lamp and the detector move around the patient. Pulse exposure, which is more often used in dentistry, helps to reduce the total radiation dose during one examination. In many aspects, cone beam tomography allows overcoming the limitations of classic CT [27,28]. It allows, depending on the need, imaging of the entire head or only a small part of the masticatory system. A single dose of X-rays concentrated in cone beams is up to 15 times lower than in the case of regular CT, which is important when frequent irradiation is necessary to assess the progress of treatment, especially in young people, and importantly, the use of low-dose CBCT protocols does not affect the quality of the obtained images [40]. There are also protocols for the use of ultra-low-dose CBCT, e.g., for pre- and postoperative assessment of alveolar cleft in children, to protect them from the negative effects of frequent exposure to X-rays [75,76,77]. CBCT also allows viewing two-dimensional images simultaneously in the sagittal, oblique, or coronal plane [61]. Other advantages of CBCT include the following: significantly lower cost, significantly shortened examination time resulting from the possibility of obtaining a qualitative image during one rotation of the device using head stabilizers, and the possibility of multiple reproduction of images on different devices [24]. In endodontics, CBCT using a limited field of view (FOV) is used in the diagnosis of periapical pathologies and imaging of root canals, in implantology for planning implant placement, assessment of peri-implant fenestration and dehiscence, in periodontology for the assessment of periodontal structures, and in maxillofacial surgery for assessment of the pathology of the third molars of the mandible and temporomandibular joints [78]. Another important application is the imaging of the airway patency, the obstruction of which may be one of the causes of some malocclusions and adenoidal facial defects, as well as sleep disorders and potentially dangerous obstructive sleep apnea [79]. Cone beam tomography, as one of the most important imaging methods in orthodontics and dentistry, is a branch of diagnostics that is worth and should be developed (Figure 5).

6. The Use of CBCT in Orthodontics in the Pediatric Population

A child’s body, whose cells are undergoing dynamic divisions, is more vulnerable than an adult to the formation of mutations under X-rays and thus to the potential for the development of tumors. Therefore, it is important to limit exposure to this type of irradiation to the minimum necessary. In the pediatric population, the so-called “ALARA principle” (As Low As Reasonably Achievable) applies, which refers to ensuring that the patient benefits more from the treatment than the potential risk of developing cancer. It includes the highest quality, evaluation of techniques and equipment, and the lowest radiation doses that will provide the necessary diagnostic information [80]. CBCT appears to be an appropriate imaging modality for children and certainly safer in terms of the amount of radiation than standard CT, also considering the much shorter duration of the examination. However, it should be noted that the amount of radiation the patient receives is several times greater than that of a standard 2D image. Moreover, even the short examination time (about 20 s) can be difficult for a child to endure in complete stillness. Even a small twitch of the head, in turn, can result in artifacts on the image and thus reduce its diagnostic value [81,82]. Some centers are developing new systems to correct and compensate for motion artifacts during the examination, but these technologies are not widely available. Instead, most offices use various types of head stabilizers, which go some way toward minimizing unwanted patient movements. The position in which the examination is performed is also important; it can be standing, lying down, or sitting. According to a study [81] involving 40 children, they tolerated CBCT best when performed in the supine position, with a foam headrest as head support. The younger the age, the higher the risk of moving was. CBCT is used in the diagnosis of many conditions in pediatric orthodontics. Some of the most common cases evaluated include an ectopic maxillary canine, retained second premolars, dental root resorption, cleft palate, the effects of trauma in the orofacial region, or supernumerary teeth. CBCT also plays a key role in preparing for surgery, tooth extraction, implant placement, and orthodontic treatment planning, as well as monitoring the course of applied therapies [80,81,82,83]. Specific studies using CBCT as a 3D imaging method before and during orthodontic treatment in children are cited below. A study by Tsolakis A.I. et al. [84] compared the results of imaging unilateral or bilateral retained maxillary canines in 20 patients using periapical, occlusal, panoramic, and CBCT X-rays. It turned out that by analyzing CBCT images, orthodontic specialists made the same diagnoses regarding the position of retained canines and the presence of resorption of adjacent dental roots. Using conventional radiographs, these descriptions differed. This study shows the advantage of CBCT in objectifying descriptions of malocclusion, in this case retained canines, which is crucial for successful orthodontic treatment. In another study, Goodell K.B. et al. [85] conducted a comparison of treatment plans for external cervical resorption (ECR) developed from periapical radiographs (PA) and CBCT. Concordance between evaluating specialists was uniformly higher for CBCT imaging, and treatment plans developed from CBCT scans differed from those created from PA radiographs in 56.7% of cases. These results also indicate the superiority of CBCT over PA when evaluating external cervical resorption. On the other hand, a study by Celebi A. et al. [86] used CBCT to determine differences in central incisor root volumes in patients with unilateral cleft lip and palate (UCLP) between the side with cleft and the healthy side, as well as compared to patients forming a control group without UCLP. The results indicated a 12.15% reduction in central incisor root volume on the cleft side compared to the side without cleft. In addition, it was found that the development of the central incisor root was significantly more dependent on the cleft (which had a smaller volume) in the female gender. The data obtained may help in planning orthodontic treatment for patients with UCLP. Another paper by Capar I.D. et al. [87] compared the effectiveness in evaluating “dens invaginatus (DI)” (a developmental anomaly where enamel invaginates into dentin) using CBCT and panoramic images rendered from CBCT images. As a result, it was found that analysis of panoramic images detected DI in 3% and CBCT images in 10.7% of patients. These results demonstrate the greater usefulness of CBCT in diagnosing DI compared to panoramic images, due to its accurate representation of the external and internal anatomy of the tooth. In the last cited study by Illipronti-Filho E. et al. [88], 20 CBCT images were taken in pediatric patients with unilateral posterior crossbite without functional mandibular displacement and with transverse jaw loss. The mandibular condyles on the right and left sides and between the crossed and uncrossed sides in sagittal and frontal views were then measured. The precise images obtained with CBCT indicated that there were no significant differences in the described dimensions, in contrast to indications in the literature suggesting an association between the size of the mandibular condyles and the presence of unilateral posterior crossbite in children.

7. Application of AI Artificial Intelligence in Cone Beam Tomography

Currently, work is underway in many centers around the world to improve cone tomography technology to expand its range of applications and improve the quality of imaging. At the same time, the use of artificial intelligence is gaining importance in many medical fields, including convolutional neural networks (CNN) and models based on deep learning [89]. The term artificial intelligence (AI) was coined in 1995 and refers to the ability of machines to perform tasks, described as intelligent, but the technology itself has been developed since the mid-20th century [90]. The AI models that have been developed present characteristics that we attribute to human intelligence, such as language comprehension, learning, reasoning, and problem solving [91]. The first AI models were based on machine learning algorithms that analyzed data sets using statistical and probabilistic methods and gradually acquired the skills to make predictions, identify new patterns, and classify the input. In addition, each new version of AI, through self-improving algorithms, became smarter than the previous one, based on a growing database of data and results, aiming to minimize the mistakes made. At the same time, there was a desire to reduce human participation in the learning process in order to achieve as much machine autonomy as possible [90,91]. The next step was the creation of deep learning (DL) models, which are able to extract a given, desired feature from the input information while learning to rely on the acquired skill in future tasks. Previously, human preprocessing was required for such data extraction [92]. DL is based on artificial neural networks (ANNs), which are a collection of interconnected units that resemble neuronal networks in the brain. Input information passes through many hidden layers of interconnected “neurons” to the output layer. Throughout the process, the parameters of the processed information are optimized with the minimization of errors between input and output data. In subsequent modifications, convolutional neural networks (CNNs) were created, in which hidden layers were replaced by convolutional, interconnecting, and fully connected layers. Convolutional layers allow narrowing the input data and reducing the amount of computation, while fully connected layers enable the optimization of information from multiple levels [93]. The latter technology is particularly effective in image analysis, including segmenting structures and detailing anomalies. Utilizing AI’s skills, its three main branches in radiology are distinguished, i.e., operational AI, which is designed to improve healthcare services; diagnostic AI, which helps interpret imaging results (X-ray, MRI, and CT); and predictive AI, which predicts treatment outcomes to some extent [90]. The development of artificial intelligence in orthodontics can help in the efficient assessment of CBCT images, saving clinicians’ time during, among others, preparation for implantological procedures [31,32,94,95]. In dental–maxillofacial radiology (DMFR), plans to use AI mainly concern automatic diagnosis of dental and maxillofacial diseases, facilitating the location of anatomical landmarks for planning orthodontic and orthognathic treatment, and overall improvement of image quality [96]. It may prove useful for automatic analysis of the upper respiratory tract, be used in oncology to individually assess the response to treatment and toxicity of radiotherapy in head and neck cancer, and be used to monitor and evaluate the effectiveness of treatment under adaptive radiotherapy (ART) [97,98,99,100,101,102]. CNNs based on deep learning methods have been proven useful for detecting dental caries on periapical radiographs, as well as classifying retained supernumerary teeth in patients with fully erupted maxillary permanent incisors on panoramic radiographs [91]. Visualization of the effects of orthodontic treatment, including facial appearance and age-related changes, is also an interesting application of AI. The potential of robotic technology for dental and maxillofacial surgery should also be mentioned [91,103]. Moreover, AI can be used in the diagnosis of osteoarthritis of the temporomandibular joint, to assess pathology in the mandibular condyle, or to detect mycosis, inflammation, and other changes in the paranasal sinuses [104,105,106]. In a study conducted by Albitar L. et al. [107], the effectiveness of AI in automatically detecting and segmenting non-obstructive mesial buccal 2 (MB2) canals on endodontically obstructive maxillary molars in cone-beam tomography images was assessed. Great potential of AI in this area has been found, but it is limited by factors such as the presence of metallic artifacts or differences in canal calcifications. In an article by Duman Ş.B. et al. [108], the diagnostic performance of an AI program based on the convolutional neural network method for the morphological classification of sella turcica (saddle turcica) on CBCT images was evaluated. The obtained results confirm the effectiveness and speed of AI in detecting anatomical landmarks, which allows for significant time savings compared to manual analysis. In the work of Gerhardt M.D.N. et al. [109], they evaluated the effectiveness of AI in marking teeth and edentulous areas in CBCT images. The obtained results indicate that the median time for indicating the desired elements by AI was 1.5 s and by humans 98 s, which clearly indicates the advantage of artificial intelligence in this respect. In turn, the article by Sin Ç. et al. [110] confirmed the effectiveness and speed of the AI program in assessing the volume of the airway at the pharyngeal level by assessing 306 CBCT images. In another study, Fontenele R.C. et al. [111] assessed the usefulness of AI in segmenting teeth depending on the type of filling on 74 CBCT images. Satisfactory results were obtained, indicating high accuracy and effectiveness of AI regardless of the material filling the tooth, which may directly facilitate the work of dentists in assessing tomography results. Vinayahalingam S. et al. [112] checked the usefulness of AI in segmenting the temporomandibular joint and mandibular fossa on CBCT images. The results indicated a much shorter time needed for analysis in the case of AI than when measurements were made by radiologists. Therefore, AI can help in the rapid evaluation of data for TMJ reconstruction. In another study, Mangano F.G. et al. [106] presented a promising concept of using AI and AR (augmented reality) for planning the implantation of dental implants in 3D technology. In an article by Jiang Y. et al. [89], they used deep learning models to segment and assess the volume and structure of the masseter muscle based on CBCT images. The study showed that this method allows for similar performance to the evaluation of classic CT images by experts and significantly exceeds the efficiency of evaluation of CBCT images by the human eye, while reducing the work time by 322 times. In the last cited study by Tao T. et al. [113], AI was used to assess the thickness of the palatal bones and soft tissues of the palate in order to select an appropriate place for the installation of an orthodontic mini-implant. As expected, the measurement results obtained using artificial intelligence were consistent with the actual situation and precise, thanks to which this technology has the potential to qualify the patient and prepare dentists for the procedure of implanting palatal mini-implants. Based on the examples cited, it can be concluded that the use of AI combined with cone tomography in dentistry and orthodontics has great potential to improve the work of clinicians and guarantee the highest quality of diagnostics. Automatic measurement and analysis of cone tomography images by AI allows, above all, to save time and effort for radiologists and dentists while guaranteeing the reliability of the results. However, this technology is new, and more clinical trials are needed to thoroughly assess its capabilities and usefulness (

Table 3. List of mentioned authors and works related to CBCT and AI.

Reference Number in Our Work	Authors	Article Title	Publication Year	Type of Work
[81]	Spin-Neto R. et al.	“Head motion and perception of discomfort by young children during simulated CBCT examinations.”	2021	Randomized controlled trial
[84]	Tsolakis A.I. et al.	“Reliability of different radiographic methods for the localization of displaced maxillary canines.”	2018	Clinical evaluation
[85]	Goodell K.B. et al.	“Impact of Cone-beam Computed Tomography on Treatment Planning for External Cervical Resorption and a Novel Axial Slice-based Classification System.”	2018	Clinical evaluation
[86]	Celebi A. et al.	“Effects of cleft lip and palate on the development of permanent upper central incisors: a cone-beam computed tomography study.”	2015	Clinical evaluation
[87]	Capar I.D. et al.	“A retrospective comparative study of cone-beam computed tomography versus rendered panoramic images in identifying the presence, types, and characteristics of dens invaginatus in a Turkish population.”	2015	Comparative study
[88]	Illipronti-Filho E. et al.	“Evaluation of mandibular condyles in children with unilateral posterior crossbite.”	2015	Prospective study
[107]	Albitar L. et al.	“Artificial Intelligence (AI) for Detection and Localization of Unobturated Second Mesial Buccal (MB2) Canals in Cone-Beam Computed Tomography (CBCT).”	2022	Validation study
[108]	Duman Ş.B. et al.	“Convolutional Neural Network Performance for Sella Turcica Segmentation and Classification Using CBCT Images.”	2022	Validation study
[109]	Gerhardt M.D.N. et al.	“Automated detection and labelling of teeth and small edentulous regions on cone-beam computed tomography using convolutional neural networks.”	2022	Validation study
[110]	Çağla S. et al.	“A deep learning algorithm proposal to automatic pharyngeal airway detection and segmentation on CBCT images”	2021	Validation study
[111]	Fontenele R.C. et al.	“Influence of dental fillings and tooth type on the performance of a novel artificial intelligence-driven tool for automatic tooth segmentation on CBCT images—A validation study”	2022	Validation study
[112]	Vinayahalingam S. et al.	“Deep learning for automated segmentation of the temporomandibular joint”	2023	Validation study
[113]	Tao T. et al.	“Artificial intelligence-assisted determination of available sites for palatal orthodontic mini implants based on palatal thickness through CBCT”	2023	Randomized controlled trial

).

8. Conclusions

The use of temporary TSAD anchoring devices has a similar meaning, which, being attached to the bone, provides stable anchoring for the appliance, at the same time enabling the minimization of the orthodontic structures installed and significantly limiting the impact on the position of the posterior teeth during the retraction of the incisors and canines. Another key issue for progress in orthodontics is the development of 3D imaging techniques, primarily cone beam tomography. The use of ultra-low-dose X-ray protocols appears to be very beneficial for the pediatric population, which is most at risk of long-term negative effects of frequent radiation exposure. At the same time, the examination produces high-quality images, without which it is currently difficult to imagine planning orthodontic treatment. The very process of analyzing the results of imaging tests, segmenting structures, and performing precise measurements of craniofacial elements is tedious and takes a lot of time for clinicians. To solve this problem, engaging artificial intelligence has become an extremely interesting trend. Many studies and analyses indicate that AI can perform tasks with the same precision with much greater efficiency and in a much shorter time. This is important not only when planning the correction of malocclusions or preparation for implantation but also in the assessment of the upper respiratory tract, whose impaired patency may, over time, lead to changes in the facial skeleton and may also cause obstructive sleep apnea. The use of self-learning technologies seems very promising, and their potential has not yet been fully explored and documented, but this will probably happen in the near future.