Three-dimensional digital applications for implant space planning in orthodontics: A narrative review

Abstract

In the digital dentistry era, new tools, algorithms, data science approaches, and computer applications are available to researchers and clinicians. However, there is also a strong need for better knowledge and understanding of multisource data applications, including three-dimensional imaging information such as cone-beam computed tomography images and digital dental models for multidisciplinary cases. In addition, artificial intelligence models and automated clinical decision systems are rising. The clinician needs to plan the treatment based on state-of-the-art diagnosis for better and more personalized treatment. This article aimed to review basic concepts and the current panorama of digital implant planning in orthodontics, with open-source and closed-source tools for assessing cone-beam computed images and digital dental models. The visualization and processing of the three-dimensional data allow better implant planning based on bone conditions, adjacent teeth and root positions, and the prognosis of the case. We showed that many tools for assessment, segmentation, and visualization of cone-beam computed tomographic images and digital dental models could facilitate the treatment planning of patients needing implants or space closure. The tools and approaches presented are toward personalized treatment and better prognosis, following the path to a more automated clinical decision system based on multisource three-dimensional data, artificial intelligence models, and digital planning. In summary, the orthodontist needs to analyze each patient individually and use different software or tools that better fit their practice, allowing efficient treatment planning and satisfactory results with an adequate prognosis.

1. Introduction

The combined orthodontic and prosthodontic therapy is a need that has always existed in clinical practice. The increasing number of adults seeking orthodontic treatment with periodontally compromised and missing tooth makes the combined treatment and synergism between the professionals a real need [1]. Therefore, it is important to not only diagnose the edentulous space but also to plan treatment based on the best evidence available and patients’ conditions. Two major approaches can be considered when planning for an edentulous space: 1) closing the space and 2) opening space for the implant and crown placement.

Treatment choice depends on the malocclusion, skeletal facial type, bone conditions, adjacent teeth conditions, and the patients’ choice. For implant restoration, clinicians should consider stress-reducing options, including a shorter cantilever, fewer offset loads to the buccal or lingual, number of implants, increased diameter of the implants, splint implant together, and optimal bone quality and quantity [2]. Ideally, all cases require a clear visualization of the results before the surgery is performed because failure of implants occurs in the range of 0% to 20% and is related to bone volume, density, and loading distribution [3].

For the reasons mentioned above, a complete assessment of the patient’s condition is necessary. Before, only two-dimensional examinations such as panoramic images (Fig. 1A) were available to clinicians, limiting the capacity of the evaluation to only vertical and horizontal dimensions of the space while evaluating the edentulous area and planning implants [3]. However, with the advances in the engineering field and computational analysis, cone-beam computed tomography (Fig. 1B) and digital dental models (Fig. 1C) are the gold-standard imaging examinations for most patients undergoing orthodontics treatment combined with restorative approaches [4]. We are approaching big data era in dental medicine with the advances in mathematics, storage capacity, and data science fields [5,6]. The amount of information available has increased significantly in the last years, requiring powerful algorithms to process data and predict treatment, diagnosis, or prognosis. In the orthodontics and restorative field, researchers have been testing different machine learning models to segment cone beam computed tomography (CBCT) images for anatomical structures assessment, helping in the treatment decision making, based on previously treated patients and data [7–10].

Fig. 1.

Example of patient’s digital records. (A) Panoramic image; (B) Cone beam computed tomography 3D rendering; (C) Digital dental model from an intraoral scanner.

This article aimed to provide insights into three-dimensional (3D) digital planning of implant placements, anatomical image segmentation, and visualization of CBCT images and digital dental models based on a narrative review of the current literature. We also highlighted data science and artificial intelligence (AI) models to help clinicians and researchers better make decisions.

2. Image modalities and steps for processing

The images of choice for precise orthodontic and prosthodontic therapy are CBCT images and 3D digital dental models (intraoral scan). The first allows the clinicians to assess the 3D spatial position of adjacent dentition, including roots positions, bone quantity and quality, management of the space available, and creation of surgical guides for implant placement. In contrast, a digital dental model is a precise tool to assess the teeth’ surface with accuracy because x-rays produce image noise on the enamel region. For didactic purposes, the basic steps of a fully digital management of orthodontic and implant treatments can be divided into five, as follows:

1. Data acquisition: the digital dental model of the upper and lower dentition needs to be acquired either with an intraoral scanner or by digitalizing the plaster model; after the digitalization, the file in the format.stl (stereolithography) needs to be stored. The CBCT scan must also be acquired for the region of interest and the Dicom files stored.

2. Data and software processing: the .stl models and Dicom files must be imported into the software of choice, such as the 3Shape Implant Studio, Blue Sky Plan, 3D Slicer, and others.

3. Data integration and planning: the two different imaging modalities need to be registered on each other using semiautomated or automated tools provided by the software; also, the bone can be segmented at this stage for visualization of the anatomy.

4. Selection of implant: at this stage, based on the integrated 3D image, bone available, anatomy, and orthodontic planning, the more suitable implant size and brand can be selected.

5. Surgical guide: at this stage, the surgical guide can be created and exported as .stl for future impressions into a 3D printer, which can be used during implant surgery.

2.1. CBCT image analysis: Visualization and planning of the edentulous space

For optimal use and measurement of x-ray–based imaging examinations, such as the CBCT, the inspection and assessment of each image slice (axial, coronal, and sagittal) are important, as seen in Fig. 2A. The 3D visualization clarifies the anatomical site conditions and adjacent teeth (Fig. 2B). Fig. 2 also shows the importance of visualization in cross-sectional views and 3D rendering. The cross-sectional inspection allows us to obtain precise information on the height and width of the future implant space and the proximity to the maxillary sinus. In contrast, the direct 3D rendering gives a general perspective of the space available. But because of the technical limitation in the automatic rendering based on picking an arbitrary threshold of voxel values, below or above which all gray values are excluded. Therefore, it is commonly used for images with the highest density values within a particular thickness, such as impacted teeth [11]. This is also illustrated in Fig. 1B, where the imaging methods such as rapid direct 3D rendering allow the assessment of the location of the impacted teeth and mandibular dentition evaluation more accurately than the panoramic x-ray (Fig. 1A).

Fig. 2.

CBCT examination of a patient in preparation for implant placement in the first premolar region. (A) Sagittal slices show the extension of the implant area. (B) 3D rendering showing the maxilla, mandible, and adjacent teeth using Invivo software (V 6.5.0).

2.2. Digital dental models and CBCT for assessment of roots and crown position—3D Digital Integration for implant planning

The use of intraoral and dental cast scanners is a routine in most practices for patients undergoing multidisciplinary treatment. The accuracy and reliability of using digital models are comparable to plaster models, with higher precision for measuring dental changes, arch changes, and available space [12,13]. Besides the bone conditions that need to be assessed, the position of the adjacent teeth roots for proper space management and implant planning are essential. Unfortunately, digital dental models cannot accurately provide this information because it only shows the crown position and gingival margins. However, our group has demonstrated that it is possible to calculate tooth crown/root movements with the 3D dental model [14]. Still, the implant surgery would need the exact position of the long axis to assess which type of implant should be chosen.

Fig. 3 shows an automated algorithm created using tools under development by the Dental and Craniofacial Bionetwork for Image Analysis - DCBIA laboratory [15,16]. The figure shows the segmentation of the roots based on AI approaches and crown segmentation from digital dental models. Still, the proper implant surgery would require integrating root information, crown, bone, and soft tissue conditions. Fig. 4 shows the 3D reconstructions of the bone, gum information, and teeth (crown and roots) in a fully integrated approach; Fig. 4A shows the assessment of the available space, and Fig. 4B shows the available area for the implant placement simulation.

Fig. 3.

3D visualization of the crown and roots position using a digital dental model and CBCT segmentation images.

Fig. 4.

Multisource images for implant planning using CBCT and digital dental models. (A) Visualization of the space available and adjacent teeth long axis. (B) Implant simulation.

The advantages of using this fusion technique are to have a precise 3D model of the teeth crowns and space available in the arch, because with a CBCT image, because of the high density of the enamel and presence of artifacts such as metal restorations, the visualization, and rendering of the dentition is limited, not allowing an accurate implant planning of tooth movement. Current AI methods have tried to improve this limitation; however, there is still a need for improvement [17]. The integration and merging of CBCT with digital dental models is an approach that has been investigated and proven to be successful [18,19]. In addition, this approach allows for overcoming the challenges of proper segmentation of the crowns using information from the 3D dental models and not from CBCT images. In summary, Fig. 5 shows what would be an ideal assessment and workflow for implant surgery and orthodontic planning.

Fig. 5.

Digital 3D workflow for implant surgery simulation.

2.3. AI approaches and clinical decision systems

AI has become an important tool in dentistry. Because of the amount of clinical and imaging data that clinicians and researchers have available and the implementation of better data science approaches, the decision-making process can be facilitated by trained algorithms. In addition, clinical decision support systems have been incorporating knowledge with patient-specific data to serve clinicians with tools that enhance this process [7,20]. Pareek et al. [21] reported that almost 4000 dental implants are marketed worldwide, with varying treatment techniques and structures. Therefore, knowing which one is better applicable to a specific patient, based on their condition, is primordial for success. In this field, AI can help the computer-aided design/computer-aided manufacturing, and panoramic radiographs classify the implant structure, and the AI approaches can help the dentist recognize and rank the implants, avoiding complications.

Researchers have also focused on detecting dental implant failures and fractures using AI methods. In 2020, Lee et al. [22] evaluated the reliability of three deep convolutional neural networks (VGGNet-19, GoogLeNet Inception-v3, and automated DCNN) for the classification of fractured dental implants in panoramic and periapical rays. They used a database with 21,398 fractured implants and found that the AI approaches had acceptable accuracy in the detection and classification of fractured implants, with the best performance obtained while using periapical radiographic images alone (automated DCNN with an area under the curve of 0.984).

AI can also assess the shape of the definitive monolithic zirconia restorations because they cannot be retouched in the mouth. Therefore, Lerner et al. [10] have demonstrated digital dental models and machine learning in fixed implant prosthodontics. The author used a fully digital protocol using AI, which allowed the successful restoration of single locking-taper implants with monolithic zirconia restorations, and they stated that the marginal adaptation, quality of interproximal and occlusal contacts, and aesthetic integration were excellent.

3. Clinical cases and clinical applications

3.1. Case 1—Guided implant surgery in a prosthodontics treatment planning

Fig. 6 shows a prosthodontics case and treatment planning with a digital setup (Integrating Digital Surface Scanners and CBCT Images) for a patient where the inferior lower right incisor is compromised because of periodontal health. In this report, the software 3Shape Implant Studio software (3Shape) was used to integrate CBCT images and digital dental models to create the implant surgical guide. This software allows the clinician to incorporate the CBCT image and intra oral-scan for decision making based on the quantity of bone available and prosthetic space available for the future crown. The first step is to perform an intraoral scanner of the patient to obtain the digital dental model and take a CBCT examination, having access to the Dicom files. Then, the 3D.stls (stereolithography) models of the upper and lower dentition and the Dicom files (CBCT) are imported into the software. A semi-automatic approach allows the registration of both image modalities, and a library containing prosthetic component information can be used to select the more accurate implant. Ultimately, the surgical guide can be virtually fabricated and sent to 3D impression. Fig. 7 shows the final implant and components using the patient dentition data (.stl models) and bone anatomy (CBCT image).

Fig. 6.

Patient data imported to the software for implant planning (3shape implant studio). (A) Visualization of the CBCT reconstruction; (B) creation of the surgical guide.

Fig. 7.

Implant components were planned into the 3shape implant studio. Design of the implant based on the anatomy of the bone and dentition.

3.2. Case 2—Orthodontic and restorative planning: minimizing the number of implants using optimal biomechanics and digital diagnosis

Figs. 8 and 9 illustrate a patient that had orthodontic treatment and implant surgery at the University of the Pacific—Orthodontics Department. The 38-year-old female patient was diagnosed with canine Class l on the right side, full cusp Class II on the left, lower midline deviation to the left, moderate to severe crowding in the upper and lower arches, and missing lower first molars bilateral, upper right first and second premolar, and upper left first premolar. Her chief complaint was the edentulous spaces, and her general dentist recommended an orthodontic consultation before her restorative treatment planning. The initial lateral cephalogram and panoramic image were generated from the CBCT and are shown in Fig. 10. The treatment proposed was to reduce the number of implants by retracting the laterals and canines into the first premolar space with moderate anchorage, maintaining the right maxillary second premolar, upright the mandibular second and third molars, close the first molar spaces, and achieving a full cusp Cl II molar and Cl I canine relationship bilaterally. This proposed plan allows the patient to have only a single implant in the first right upper premolar region. The pre-adjusted 0.022 × 0.028” MBT prescription fixed appliance (Victory, 3M Unitek) was used. Fig. 9A shows the progress in the orthodontic treatment. TADs (1.4 mm diameter, 6 mm length, Vector, Ormco) were used to minimize the side effects of the continuous archwire in the lower dentition that was used to upright the lower molars and to close the space without losing anterior anchorage, and gable bends were added to promote mesial root tip of the second molars in the 0.017 × 0.025” SS archwire. In the upper arch, the sequence progressed from 0.016 × 0.022” NiTi to 0.019 × 0.025” NiTi to 0.019 × 0.025” TMA. The spaces were closed by sliding mechanics except for the upper right second premolar, which was maintained, and the space was adequate for an implant. Fig. 9B shows the final results after single implant placement, and Fig. 11 shows the lateral cephalogram, panoramic image, and cephalometric tracing immediately after treatment.

Fig. 8.

CBCT 3D rendering of a patient needing an implant. The image allows the visualization of the treatment choice and planning of the biomechanics.

Fig. 9.

Clinical case showing a patient that had orthodontic combined with surgery for the implant of the upper right premolar. (A) Initial, progress, and final photos of the orthodontics treatment. Most spaces were closed, and only one implant was planned. (B) Implant x-rays and final photos after the implant surgery.

Fig. 10.

Initial treatment records. (A) Pretreatment lateral cephalogram, (B) lateral cephalometric tracing, and (C) panoramic image.

Fig. 11.

Final treatment records. (A) Final treatment lateral cephalogram, (B) lateral cephalometric tracing, and (C) panoramic image.

3.3. Case 3—Orthodontic planning using digital tools for orthodontics and implant space treatment simulation

A 17-year-old male patient presented to the University of the Pacific - Orthodontics clinic with the chief complaint of retained deciduous mandibular and maxillary molars and impacted upper second premolars. In addition, he had the absence of the lower second premolar, requiring detailed treatment planning to address his condition (Fig. 12A). Therefore, a virtual setup was performed using the software Archform and ClinCheck. Fig. 12B shows his initial digital models, and Fig. 12C shows his CBCT images for assessing the conditions of the lower deciduous molar. Because the patient’s guardian refuses to extract the second upper premolar, the two main treatments proposed were as follows: 1) orthodontic traction of the impacted upper second premolars and maintenance of the lower space for future implant in the deciduous molar region or 2) orthodontic premolar traction of the impacted upper second premolars and maintenance of the space for future implant in the lower deciduous molar region. Next, a virtual simulation was performed with two different software: Archform (Fig. 12E) and ClinCheck (Fig. 12D). This setup allowed us to see that the maintenance of space was an appropriate choice because of the occlusion of the second upper molar.

Fig. 12.

Simulation of treatment in a patient with the presence of bilateral molar deciduous and congenital absence of bilateral lower second premolars. (A) Intraoral photos; (B) digital models; (C) CBCT 3D rendering; (D) simulation of treatment with the closure of the spaces; and (E) simulation of treatment with preservation of the space for implant planning.

4. Discussion

The digitalization of treatments is rising in orthodontics and prosthodontics with AI approaches. Especially with the use of aligners, digital dental models, and more access to CBCT images, better and more personalized treatment planning is possible. In a recent publication, Shroff et al. [1] showed two case reports using a digital workflow with virtual orthodontic planning and the design of surgical guides for implant placement. As the main advantages of using digital aligner therapy, they cited the possibility of using a 3D setup and accurate planning of the final tooth position, simulation of orthodontic movement, and space creation for the implant; but the disadvantages are the relatively high cost compared with fixed appliances and the need for patient cooperation while using the aligners. In comparison, we also showed in the current study that the simulation of tooth movement could be done without the need for aligner therapy. Fig. 12D and 12E show a virtual setup with the purpose of treatment planning only, where the patient received fixed appliances but had the digital setup for a better prediction of the occlusion after closing the implant space (Fig. 12D) or implant space maintenance (Fig. 12E).

It is important to highlight the multidisciplinary aspects of or thodontics and implant treatment involving specialties such as periodontics, dental implant, orthodontics, and prosthodontics. Blasi et al. [23] presented the digital preorthodontic implant placement, showing the steps for a fully digitized treatment using digital bracket placement and guided surgeries for periodontics and implant purposes. In addition, Tarraf et al. [24], in 2018, pointed out that digital technology has a great impact in the medical and dental field, allowing personalized and better treatment options with computer-aided design/computer-aided manufacturing techniques, indirect bonding trays, customized wires, and even remote monitoring of treatment. The authors also showed the integration of 3D photos and facial scans in treatment planning.

Overall, the digital combined orthodontics and implant surgery is an emerging and needed tool. This approach allows for better and more robust treatment planning, with fewer variables and more predictability. This was also demonstrated by Spalthoff et al. [25], who evaluated the efficiency of a digital workflow of prosthetic teeth positioning between virtual standard-sized digitally constructed and conventional dental laboratory-fabricated prostheses. They found that the digital workflow provided accurate final results.

5. Conclusion

Combining multisource images such as CBCT and 3D digital dental models is essential for proper planning and managing implant surgery combined with orthodontic treatment. In addition, the use of data science approaches, advances in the image analysis field, and new AI approaches are becoming more popular among clinicians because of translational research and software availability. Therefore, better and more personalized treatment can be available, helping the clinical decision making and the prognosis.