A novel methodology for analysing dental implant positional changes from virtual planning to placement without CBCT

Abstract

Introduction This paper introduces a novel methodology for analysing dental implant positioning in vivo, advancing beyond traditional methods that rely on post-placement cone beam computed tomography (CBCT) scans.

Method The core of the methodology is comparing stereolithography (STL) files, representing pre-planned and actual post-placement implant positions. These STL files, exported from guided surgery planning and computer-aided design software, focus on clinically significant key points, like the apical and coronal midpoints. Additionally, the method uses pose detection, differing fundamentally from CBCT scan approximations.

Discussion Relying on pose detection instead of scanner resolution, this method aligns with international standards and overcomes CBCT and intra-oral scanner limitations. It allows for a more precise and accurate assessment of implant positions, independent of scanner technology constraints. Further refinements include potential detailed reporting of apical deviations, enhancing implant placement accuracy.

Conclusion This research has significant implications for dental implantology, enhancing implant placement precision and overall procedure success. Introducing an automated process for report generation through a batch script improves efficiency and precision, streamlining the analysis process. This innovation sets the stage for future technological advancements, improving both the accuracy and reliability of implant placement assessments.

Key points

• Improved precision: the methodology enhances the accuracy of dental implant positioning verification by comparing pre-planned and actual implant positions using stereolithography files and pose detection, rather than approximation through cone beam computed tomography (CBCT) overlaid with pre-planned position.

• Reduced radiation exposure: this approach eliminates the need for post-placement CBCT scans, significantly reducing overall patient radiation exposure and bypassing ethical considerations in taking both a pre-planned and post-surgery CBCT.

• Efficient analysis: an automated process for generating detailed results streamlines the evaluation of implant positions, promoting efficiency in clinical practice.

• Overcomes scanner limitations: using pose detection technology provides a more precise assessment of implant positions, independent of the limitations of CBCT and intra-oral scanners.

Introduction

The precise positioning of dental implants is a critical factor for the success of dental restorations. Traditionally, evaluating the accuracy of implant placement has involved comparing post-operative cone beam computed tomography (CBCT) scans with pre-operative stereolithography (STL) models. This process typically involves aligning an STL model, which represents the pre-planned implant position, with a CBCT scan obtained after the implant is placed. This alignment helps clinicians assess any deviations between the intended and actual positions of the implants.

Software tools such as CloudCompare are used to quantify these discrepancies. CloudCompare is a software that allows for the comparison of 3D models and point clouds, which are sets of data points in space produced by 3D scanners. This software enables clinicians to observe and measure differences between the planned positions (from STL models) and the actual positions (from CBCT scans) of dental implants.

However, with growing concerns about radiation exposure from repeated CBCT scans and the associated costs, recent advancements have led to the exploration of less invasive techniques. For instance, intra-oral scanners, which capture direct optical impressions inside the mouth, have been identified as a promising alternative. Jorba-García et al.¹ demonstrated that intra-oral scanners can assess implant positions with high accuracy, potentially replacing CBCT in certain clinical scenarios. Similarly, optical imaging techniques, which involve capturing visual information through light-based sensors, offer non-invasive ways to obtain precise measurements of implant positioning. Research by Sarment et al.² highlights the effectiveness of these techniques in dental implantology.

This shift towards innovative methodologies that do not rely on traditional CBCT scans underscores a broader trend in dental research. It emphasises the development of new software tools and methodologies aimed at improving the accuracy of implant placement while reducing patient exposure to radiation and lowering treatment costs. This evolving landscape in dental implantology is pivotal, as the precision of implant placement has a significant impact on the long-term success of prosthetic restorations, as evidenced by the findings of Jung et al.³

Limitations of using CBCT scan post-implant placement for comparing implant position to pre-planned virtual position

CBCT scans are a cornerstone in both the pre-operative planning and post-operative evaluation in guided dental implant surgeries. However, the use of CBCT scans to compare the actual implant position with its pre-planned virtual position presents several notable limitations, as per below.

Radiation dose

The major concern with the routine use of CBCT scans for pre- and post-implant assessments is radiation exposure. Typically, two CBCT scans - one before and one after implant placement - are required, doubling the radiation dose received by the patient. Although CBCT scans have a lower radiation output than conventional computed tomography (CT) scans, the cumulative radiation dose from multiple scans raises significant safety concerns. The need to balance patient safety with diagnostic needs has spurred research into low-radiation alternatives and protocols that minimise exposure while maintaining image quality.⁴

Artefacts and scatter

CBCT scans can be compromised by artefacts and scatter, particularly around metallic objects, such as dental implants. Metal in the surgical area can cause beam hardening and scatter, which manifests as streaks and shadows in the imaging, degrading the quality and accuracy of the scan. These artefacts make it challenging to precisely determine the post-placement position of implants, introducing errors into the analysis. Efforts to improve CBCT technology focus on reducing these distortions and enhancing image clarity.^5,6

Inherent error and approximation

The presence of artefacts and scattering inevitably leads to approximations in determining the actual position of the implant post-placement. Such approximations introduce potential errors, which can affect the clinical evaluation of how closely the implant placement aligns with the planned position. This uncertainty highlights the necessity for ongoing development of more precise imaging techniques and software that can provide more accurate assessments.⁷

In summary, while CBCT scans are invaluable for the detailed planning and subsequent analysis of dental implant placements, their limitations, including increased radiation exposure, susceptibility to artefacts, and the consequential inherent errors in position approximation, pose considerable challenges. These issues necessitate careful interpretation of CBCT data and underscore the urgent need for technological advancements to improve the accuracy and reliability of implant positioning assessments.

Novel approach

This research presents a novel methodology developed by the author for comparing the pre-planned implant position with the actual position post-implant placement. Using STL files, which contain detailed coordinates of vertices representing both the planned and actual implant positions, this technique enables a highly precise comparison. The STL file for the planned implant is generated using guided surgery planning software, which provides a virtual blueprint of the desired outcome. Conversely, the STL file for the post-placement implant is derived from computer-aided design (CAD) software, which captures the real-world outcome of the surgical procedure. This approach allows for a direct and detailed comparison of key anatomical points, notably the apical (tip) and coronal (head) midpoints of the implant, which are crucial for assessing the accuracy of implant placement.

Pose detection technologies employed in motion capture systems,⁸ as shown in Figure 1, offer a valuable framework for understanding the application of similar technologies in dental implantology to that described in the novel approach within this paper. In motion capture, precise tracking of an object's or individual's movement in three-dimensional space provides detailed insights into position and orientation changes over time. This high level of accuracy and real-time feedback is crucial for applications such as animation and biomechanics. When applied to dental implantology using the novel approach described herein, the principles of pose detection can enhance the precision with which implants are measured and placed by understanding the change of the exact same STL derived from the pre-operative guided surgery planning to the STL generated from the post-placement STL derived from the CAD software. By adopting pose detection methodologies for tracking and analysing movements using the novel software developed from this novel approach, we can more accurately compare the pre-planned and actual positions of implants, ensuring optimal alignment and improving surgical outcomes.

Fig. 1

Pose detection used in motion capture. Reproduced with permission from Vazquez et al., ‘Development of posture-specific computational phantoms using motion capture technology and application to radiation dose-reconstruction for the 1999 Tokai-Mura nuclear criticality accident', Physics in Medicine & Biology, vol 59, pp 5277-5286, 2014, DOI: 10.1088/0031-9155/59/18/5277, ©Institute of Physics and Engineering in Medicine. Reproduced by permission of IOP Publishing Ltd. All rights reserved¹³

Incorporating pose detection in implant position analysis

The core innovation of this methodology is the integration of pose detection technologies, which significantly refines the analysis of dental implant positioning. Unlike traditional methods that primarily rely on approximations from post-placement CBCT scans, pose detection evaluates the exact spatial orientation and position of an implant in three-dimensional space. This technique aligns with ISO (International Organisation for Standardisation) standards that emphasise accurate pose detection over traditional measures of scanner resolution, which often do not capture the true clinical accuracy.

Pose detection operates by analysing the relative orientation (angular disposition) and position (spatial coordinates) of the implant with respect to a predefined reference frame. This is critical in dental implantology, where even minute deviations from the planned orientation and position can affect the functional and aesthetic outcomes of the implant-supported restorations.

The robustness of this method lies in its ability to transcend the limitations imposed by scanner resolution. Traditional imaging techniques often face challenges, such as scatter and artefacts, particularly around metallic objects, which can obscure clear imaging. By focusing on the relational data between pre-planned and actual implant positions, rather than absolute resolution metrics, this method provides a more stable and reliable framework for evaluating surgical outcomes.

Deepening understanding of pose detection in dental implantology

Pose detection, as applied in dental implantology, involves complex algorithms that discern the three-dimensional orientation and location of implants. This capability is analogous to technologies used in robotics and computer vision. For example, in robotics, pose detection is essential for the operation of autonomous robots that must navigate complex environments, recognising and adapting to their surroundings in real-time.⁹ In computer vision, pose estimation algorithms are used to track the position and orientation of objects within digital images, critical for creating augmented reality experiences where digital and real-world elements interact seamlessly.¹⁰

In the context of dental implantology, pose detection provides a strategic advantage by allowing clinicians to precisely determine how closely the actual implant placement matches the virtual planning. This precise alignment is vital for ensuring that the functional load on the implant is distributed correctly, which is crucial for the long-term success of the prosthetic restoration. Moreover, by using pose detection, clinicians can significantly reduce the reliance on high-resolution scans, which are often limited by technical discrepancies and can expose patients to unnecessary radiation.

Through the application of pose detection, the methodology detailed in this paper not only addresses the limitations of traditional imaging but also paves the way for more advanced diagnostics and treatment planning in dental implantology. It promises enhanced accuracy in implant placement, minimising the risk of complications and improving clinical outcomes.

Alignment process

The alignment process is crucial to ensure a valid result with optimum alignment. The process involves ignoring the actual implant initially and aligning the post-placement scan of the edentulous arch with scan bodies, with the pre planning edentulous scan using the dental surfaces and gingivae. The alignment process was meticulously executed in several steps:

• 1. Export of the original virtual implant planning as native implant STLs, labelled according to each artificial edentulous mandible from the chosen guided surgery software

• 2. Scanning of the implant scan bodies in the bone post-implant placement procedure of the artificial jaw with a lab scanner

• 3. Utilisation of CAD software to align the scan bodies and export pure native post-placement STLs of the implants

• 4. Exportation of each individual post-placement virtual implant with maintained XYZ position

• 5. Specific nomenclature was used for each exported implant STL based on the artificial jaw number and position.

Figure 2 illustrates the overlaid STL files of the implants, showing the pre-planned versus the actual post-placement positions in a three-dimensional space. The visualisation helps in quantifying the deviation, if any, from the planned alignment, providing a clear and immediate representation of the accuracy of implant placement.

Fig. 2

The pre-planned position and post-placed position STL of the implants overlaid to see the change in position in three-dimensional space

Data measurements and analysis

Once the alignment of the implant STL files is validated using the guided surgery and CAD software, the next critical step involves measuring the deviation between the planned and the actual positions of the implants. This is done by analysing the post-placement virtual implant STLs against the pre-planned virtual STLs. The following detailed methods were employed to ensure robust and precise measurements.

Measurement techniques

Procrustes transformation

This statistical method is used to analyse the shape of objects by eliminating variations in size, position and orientation. In the context of this study, Procrustes transformation was considered to standardise the implant positions before comparing deviations, ensuring that differences are purely due to misalignment rather than other geometric variations.

Matrix decomposition

Decomposing the transformation matrix into translation and rotation components helps in understanding how much of the deviation is due to linear displacement (translation) and how much is due to angular displacement (rotation). This distinction is crucial for assessing the precision of implant placement.

Linear measurement from clinically relevant key points

This involves direct measurements of distances between corresponding key points (such as the apical and coronal midpoints) on the pre-planned and actual implant meshes. This method provides a straightforward, interpretable metric of deviation that is highly relevant clinically.

Data recording and statistical analysis

The results of the deviations were meticulously recorded into a table format, which can be easily imported into statistical software like Excel or SPSS for further analysis. Statistical techniques, such as mean deviation, standard deviation and error propagation, were used to assess the overall accuracy of the implant placement process and identify any consistent patterns or outliers in the data.

Software tools developed for analysis

A custom-made C++ program developed by Andrew Keeling at Leeds University¹¹ was used for calculating positional changes in the XYZ coordinates of the STLs. This program applies complex algorithms to precisely quantify the deviations between the planned and actual implant positions.

Additionally, a custom-made graphical user interface, programed by the author of this paper and made available through the International Digital Dental Academy,¹² was used to facilitate the calculation and visualisation of these deviations. This interface allows users to interactively explore the data and understand the nature of the positional changes more intuitively.

Figure 3 provides a visualisation of the custom software used to calculate the deviations in implant positions, highlighting the sophisticated algorithms and computational methods employed. Figure 4 shows the graphical user interface that assists in the analysis of implant positional changes. It is designed to be user-friendly and accessible, promoting wider use and understanding of the analysis process.

Fig. 3

The positional change programming created at Leeds University¹¹

Fig. 4

The positional change calculator software graphical interface

Challenges and solutions

One of the challenges encountered was the inconsistency in the ordering of the vertices in STL files. One issue with STL files in general is that they sometimes jumble the ordering of the vertices, meaning that although you have the same number of triangles and vertices in both files, the vertex IDs are different.

Instead, the code does the following:

Measures the angle between the long axes of the two implants. (This is important because we now ignore rotational angle errors caused simply by the screw thread)

Calculates two points, A and B, which sit centrally and at either end of the implant

Measures the distance between point A on pre to point A on post, and similarly for point B

Saves both file names, the A-to-A distance, the XYZ coordinate of point A on the post-implant, the B-to-B distance, the XYZ coordinate of point B on the post implant, and finally, the angle between long axes in degrees.

This method allowed for the identification of apical and coronal ends and the measurement of key outcomes, like apical deviation and angular deviation.

Discussion

Implications and possible uses

The methodology developed to overcome the vertex ordering challenge in STL files has several important implications, as below.

Enhanced accuracy in implant placement

By focusing on anatomically significant points and ignoring non-clinical discrepancies, the technique ensures a high level of accuracy in evaluating the fidelity of implant placement relative to the surgical plan. By providing a more exact comparison between pre-planned and actual implant positions, this method helps to refine surgical techniques, ensuring that implants are placed more accurately according to the surgical plan.

Improved clinical outcomes

Accurate measurements of deviations could potentially help in assessing and potentially correcting the orientation and position of implants, leading to better functional and aesthetic outcomes. The increased precision in implant placement directly contributes to the overall success rate of dental implant procedures, potentially reducing the incidence of complications associated with misalignment.

Application in quality control

This method can be employed as a standard practice in quality control during implant surgeries, ensuring that each implant is placed as accurately as possible according to the surgical plan. The ability to precisely compare planned and actual implant positions offers valuable insights into the effectiveness of surgical planning and the execution, enabling clinicians to make informed adjustments to their practice.

Research and development

The detailed data collected can be used in research to further understand the factors influencing implant placement accuracy and to develop new technologies or techniques that enhance the precision of surgical outcomes. Expanding the reporting to include details about apical deviations relative to the long axis of the post-implant could provide deeper insights into both the extent and direction of the implant apex's deviation in general guided implant surgery. This level of detail is crucial for comprehensively understanding implant placement nuances and could potentially lead to improved surgical outcomes.

This research introduces a cutting-edge approach for analysing dental implant positioning in in vitro mandible studies, marking a significant advancement from traditional methodologies. Traditional methods typically rely on post-placement CBCT scans which approximate implant positions. In contrast, this innovative technique focuses on clinically relevant key points for more accurate and precise comparisons, effectively overcoming the limitations associated with conventional imaging methods.

Future directions

Looking forward, the potential for technological advancements in this domain is vast.

Integration with advanced imaging technologies

Incorporating this methodology with newer, more advanced imaging technologies could further refine the precision of implant placement analyses. This approach has the potential to be integrated into advanced surgical planning and simulation software, providing real-time feedback to surgeons and enhancing the predictability of implant surgeries. Moreover, it could be adapted for use in training programs, allowing dental students and new surgeons to visualise and understand the importance of precision in implant placement.

Automated reporting processes

The development of an automated process for generating detailed reports of the results is another area of potential advancement. Implementing a batch script for automating the creation of result text files could exemplify a move towards greater efficiency and precision in dental implantology research.

Development of real-time analysis tools

Creating tools that provide real-time feedback during surgeries could revolutionise surgical practices, making dental implant procedures more reliable and successful.

Future innovations

This study sets the stage for future innovations that could significantly enhance the efficacy and reliability of dental implantology with guided implant surgery. The move towards more accurate, precise and efficient methodologies will undoubtedly continue to propel the field forward, improving patient outcomes and shaping the future of dental surgical practices.

Conclusion

This research pioneers a novel method for analysing dental implant positioning, offering a significant improvement over traditional CBCT scan-based approximations. By emphasising clinically relevant key points, this approach allows for more precise and accurate comparisons of implant positions, enhancing the precision of placements and the overall success of procedures. While promising, the method could benefit from further enhancements, such as detailed reporting of apical deviations and the development of automated reporting processes. These improvements have the potential to refine surgical planning and execution further, setting the stage for future technological advancements that could transform dental implantology.