Integration of multimodal patient data in a full-cycle digital protocol for dental implantation and prosthetic rehabilitation

Introduction

Dental implantology has long moved beyond the boundaries of a purely surgical discipline: the question of where to place an implant has given way to the question of which restoration it should ultimately support and how the entire pathway from diagnosis to final delivery can be completed without loss of spatial information. This shift in emphasis, from the surgical stage to a planned information chain, now defines the quality of a full-cycle digital protocol in which implant position is planned according to the requirements of the future prosthetic restoration [1, 20].

At the same time, the technological armamentarium has clearly outpaced its methodological conceptualization. Cone-beam computed tomography (CBCT), intraoral scanners, guided surgery planning systems, and CAD/CAM platforms have been described in detail as stand-alone tools; howeve­r, it remains unclear how they are interconnected within a specific protocol, at which transition the risk of geometric data distortion is greatest, and how this risk is documented. Some of these questions have already been answered: the discrepancy in registering CBCT with surface scans depends on the segmentation method and the number of metal restorations [21], whereas static navigation provides clinically acceptable, although not error-free, accuracy in transferring the virtual plan [22, 23]. Systematic reviews also document the rapid adoption of artificial intelligence (AI) and digital solutions in implant planning, yet cross-system transitions are unevenly covered in the available literature [4, 15].

In particular, few clinical protocols have been described in which the integration of multimodal data is treated not as a technical procedure but as a verified sequence of transitions, each with a clearly defined checkpoint. Most published clinical reports present the treatment outcome without revealing the actual mechanism by which spatial information is transferred between stages [2, 9, 10]. This gap defined the aim of the present study.

The aim of the study is the description and documentation of a sequential full-cycle protocol for integrating multimodal patient data in dental implantation and prosthetic rehabilitation, the identification of checkpoints at each cross-system transition, and the analysis of risks associated with the accumulation of digital error along the entire chain.

Сase description

The study was designed as a case report in accordance with the CARE guidelines on the structure and completeness of reporting [24].

Baseline clinical status

A 47-year-old male patient presented to the clinic; written informed consent for publication of de-identified clinical data was obtained. The initial examination revealed partial secondary edentulism at the sites of missing teeth 36 and 46; the adjacent teeth had preserved periodontal support, and oral hygiene was satisfactory. A panoramic radiograph showed sufficient vertical bone volume in the regions of interest; however, CBCT was prescribed to assess bone density, alveolar ridge width, and the safe distance to the mandibular canal.

The general medical history revealed no contraindications to surgical treatment; the patient was a nonsmoker, and no anticoagulant use was recorded. Following a joint consultation between the implant surgeon and the prosthodontist, a decision was made to perform bilateral implant placement in the mandible, followed by fixed prosthetic rehabilitation within a full-cycle digital protocol.

Data sources and acquisition sequence

The information base of the protocol was built from four typologically distinct sources. Each source produced its own digital object format, and none of them alone provided a sufficient picture for simultaneous surgical and prosthetic decision-making; this is the fundamental feature of the multimodal approach [2, 10, 11].

CBCT was performed with a field of view sufficient to cove­r both jaws. The resulting DICOM dataset served as the basis for a three-dimensional model of the bone structures, which was used to assess anatomical landmarks, bone volume, and density and to plan the axial trajectories of the implants while maintaining the minimum permissible distances to anatomical structures.

Primary intraoral scanning (IOS) was performed by sequential surface capture and yielded two STL models and a bite registration file. These surface data captured the geomet­ry of the dental arches and the soft tissues of the alveolar ridge in the edentulous areas, that is, precisely the structures that CBCT reproduces considerably less accurately because of artifacts and limited soft-tissue contrast [1, 21].

Clinical photographic documentation (frontal, lateral, and occlusal views) recorded the baseline occlusion, periodontal tissues, and smile line. At later stages, these photographs served as an esthetic reference during CAD design of the restorations.

Postoperative intraoral scanning with scan bodies was performed 7–10 days after surgery. The resulting STL model contained the three-dimensional coordinates of each scan body, which made it possible to transfer the actual implant positions directly into the CAD environment. Given the distal location of the implants, auxiliary clips were applied to the scan bodies, a technique proposed to improve the accuracy of complete-arch digital scans [13].

Software environment and functional roles

DICOM and STL data were integrated in implant planning software that supports static navigation. The software performed three functions: it registered the surface model to the volumetric radiographic dataset, enabled prosthetically driven implant positioning, and generated a surgical guide for subsequent 3D printing. The guiding principle of positioning was backward planning: the optimal position of the future restoration was determined first, and only then the corresponding implant position, which was checked for compatibility with the bone anatomy [4, 20].

The definitive restorations were designed in exocad DentalCAD. Input data comprised the postoperative STL model with scan bodies, antagonist models, and the occlusion file. The system allowed interactive crown design with control of occlusal contacts, path of insertion, and esthetic parame­ters, after which the finalized CAD model was transferred to the production environment for milling or additive manufacturing [17, 18].

Timeline and sequence of transitions

The protocol was carried out in nine sequential stages. The first stage comprised the initial diagnostic workup, including history taking, photographic documentation, and panoramic radiography; the second involved CBCT acquisition of the DICOM dataset; and the third, primary intraoral scanning, after which the STL files were imported into the planning software and registered with the DICOM data. The fourth stage was the planning itself: placement of digital implant analogs and their verification on the radiographic and surface layers. At the fifth stage, the surgical guide was fabricated and checked on a model, and at the sixth, the implants were placed under guide control. The seventh stage involved postoperative scanning with scan bodies and generation of the final STL model; the eighth, CAD design and fabrication of the restorations; and the ninth, clinical try-in, delivery, and radiographic and photographic follow-up [2, 7–9].

Results

The full-cycle protocol was implemented without interruption, from the initial DICOM dataset to the delivered definitive restoration. At each of the eight documented transitions between data sources, a specific digital object was generated and became the input for the next step. The distribution of transitions, clinical decisions, output objects, and checkpoints is presented in Table 1, where a separate column lists the corresponding literature sources.

Table 1. Information transitions and checkpoints of the digital protocol

The first documented link was the spatial registration of DICOM and STL data. The surface data were registered to the volumetric radiographic dataset using anatomical landmarks, namely the crowns of the teeth adjacent to the edentulous sites. Registration quality was checked visually in several views; no discrepancies requiring rescanning were detected. This transition is the most vulnerable link in the protocol, because the geometric registration error is carried over to all subsequent digital objects, primarily the surgical guide [21].

The registered model yielded a verified implant plan. Positioning followed the logic of backward planning: the prosthetic requirements for the restoration (occlusion, interarch distance, and path of insertion) were defined first, and only then were these parameters matched against the anatomical capacity of the bone. All planned positions met both the prosthetic criteria and the anatomical constraints, and the digital plan was saved in a format suitable for surgical guide generation.

The next digital object was the surgical guide, fabricated by additive manufacturing. Before surgery, the guide was checked for seating stability and for alignment of the guide sleeves with the planned axes; immediately before implant placement, its seating was verified clinically, as documented in the operative record. Both implants were placed through the guide, and the operative record noted no intraoperative deviations from the plan.

After the surgical stage, a postoperative STL model with scan bodies was obtained. Scanning was performed 7–10 days after surgery; before the procedure, the connection between the scan bodies and the implants was checked clinically. Given the distal position of the implants, auxiliary clips were used to stabilize spatial registration [13]. The resulting model contained the three-dimensional coordinates required for the transition to CAD design.

On this basis, the CAD model of the definitive restorations (crowns at positions 36 and 46) was finalized in exocad DentalCAD. The design was performed interactively with verification of occlusal contacts and esthetic parameters; after model approval, the corresponding files were transferred to the production environment for milling or 3D printing [14, 17, 18].

The final outcome was the delivered definitive restoration. Radiographic control confirmed close marginal fit, occlusion was verified clinically, and photographic documentation of the final result showed satisfactory esthetic and functional outcomes. All eight checkpoints were passed sequentially, with no documented transition omitted.

Discussion

The present case report is not intended to demonstrate the superiority of digital over analog methods. Its significance lies in showing that a digital protocol acquires real clinical value only when controlled and verified transitions exist between its stages. A set of separate digital tools that are not linked by documented transfer of spatial data is formally digital but functionally no different from a collection of isolated procedures. Yet the controllability of cross-system transitions remains outside the scope of analysis in most published clinical reports [2, 9, 12].

Registration of CBCT and intraoral scan data is among the most underestimated transitions in the digital chain. Registration accuracy depends substantially on the number and quality of common landmarks shared by the volumetric and surface models, as well as on the segmentation method and artifacts from metal restorations: in the study by T. Flügge et al. involving 36 patients, the mean discrepancy between the models after registration was 0.54 mm [21]. In partial defects, this landmark role is naturally fulfilled by the remaining teeth, whereas in fully edentulous arches it is assumed by radiopaque markers or dedicated registration devices. Supplementing the registration protocol with facial data, as proposed by M. V. Valenzuela et al., expands the possibilities for esthetic and occlusal planning [10]. In the present case, photographic documentation was also part of the protocol, although it served primarily as an esthetic reference rather than as a registration landmark.

Template-based guided surgery (static navigation) is a well-studied method. According to a meta-analysis of 20 clinical studies by A. Tahmaseb et al., the mean deviation of the actual implant position from the planned one was 1.2 mm at the entry point, 1.4 mm at the apex, and 3.5° in angulation, with higher accuracy in partially than in fully edentulous patients [22]; a later meta-analysis of 67 publications confirmed comparable values [23]. Nevertheless, the guide is not a perfect means of plan transfer: errors related to guide seating instability, 3D printing quality, and operator technique persist [20]. The coronal, apical, and angular components of deviation should be distinguished, since their clinical significance varies with implant location and anatomical constraints [22, 23]. In the described protocol, planned and actual implant positions were not compared quantitatively; therefore, no numerical claims regarding the accuracy of guided implementation can be made for this case.

Transfer of implant positions via scan bodies carries its own risk point: error accumulation as the scanner is moved over long distances. The systematic review and meta-analysis by A. Pozzi et al., which included 13 studies of complete-arch reconstructions with 4–8 implants, showed higher accuracy of photogrammetry than of intraoral scanning in 10 of them [16]. For two single implants, as in the present case, intraoral scanning remains a methodologically sound choice, particularly when combined with auxiliary clips on the scan bodies [13].

The role of 3D printing in the described protocol is twofold: it is used to fabricate the surgical guide and, potentially, to produce prosthetic restorations. Review articles cover a wide range of clinical applications of additive manufacturing, from surgical guides to provisional and definitive restorations [17–19]. The review by H.A. Adel and A.A. Abuljabbar emphasizes that combining 3D printing with CAD/CAM systems and intraoral scanners reduces errors associated with manual steps and accelerates the fabrication of personalized devices, including surgical guides [19]. This is directly related to the logic of the protocol: if the guide or restoration is fabricated with a geometric error, the accuracy of the preceding planning stages loses its practical significance.

The potential of AI for future iterations of such protocols deserves separate consideration. AI systems already demonstrate the ability to segment anatomical structures on CBCT scans automatically and to predict the number and positions of implants [5, 6]; in particular, in the study by S.B. Mun et al., a deep learning model predicted the number of implants required on the basis of panoramic radiographs and patients’ clinical data [5]. Systematic reviews summarize the rapid growth in the number of such applications in implant planning [4, 15]; however, any automation of clinical decision-making adds yet another transition to the digital chain, one that likewise requires verification.

The traceability concept proposed by S. Chen et al. in the form of a platform for managing the multimodal data of implant patients can be regarded as a formalization of the same approach that the present study implements at the level of a clinical protocol [3]. The platform, which contains 715 standardized data elements, is designed not only for data storage but also for tracking the chronological sequence of data acquisition [3]. Without such an architecture, a digital pathway remains digital in name only.

As shown in Table 2, each critical risk point has a specific and practically feasible control method; therefore, provided that the protocol is followed, none of them is an inevitable source of error. This point is fundamental: the problem of digital chains in clinical practice lies not in the absence of solutions but in the unsystematic application of already known control measures.

Table 2. Critical risk points in the digital chain of the protocol

The strengths of this work include the consistent documentation of each transition between data sources in a tabular format that can be adapted to various clinical scenarios. At the same time, the work has limitations: the single-case design without a control group; the lack of a quantitative comparison of planned and actual implant positions; the description of only a partially edentulous case and a single surgical method (static navigation); and the dependence of the outcome on specific software and operator experience. Extrapolation to complete-arch reconstructions requires corresponding adaptation of the implant position transfer protocol [7, 8]. Nevertheless, the technical article by L. Lepidi et al. on the concept of the four-dimensional virtual patient [11] and the digital protocol of E. Lefrançois et al. for rehabilitation with fixed and removable prostheses [12] indicate that the principle of sequential data registration based on stable landmarks is applicable in fundamentally different clinical scenarios, which points to the methodological transferability of the presented approach.

Conclusions

1. A full-cycle digital protocol for dental implantation acquires clinical value not through the range of technologies involved but through a verified sequence of documented transitions between multimodal data sources: without control of each transition, the concept of a digital protocol loses its practical meaning.

2. Spatial registration of CBCT (DICOM) and intraoral scan (STL) data is the most methodologically vulnerable link in the chain: registration error at this step propagates to the surgical guide and the CAD model of the restoration; therefore, its verification is a mandatory requirement rather than a recommendation.

3. Transfer of implant positions via scan bodies requires clinical verification of connection quality and, for extensive reconstructions, the use of additional stabilizing elements or alternative methods such as photogrammetry, because error accumulation during complete-arch scanning is a documented systemic risk.

4. The absence of a quantitative comparison of planned and actual implant positions is the main limitation of the described protocol; incorporating such a comparison as a mandatory checkpoint defines a priority direction for further research.

Informed consent

The patient provided written informed consent for publication of de-identified clinical data.

Conflict of interest

The author declares no conflict of interest.

Funding

This study received no external funding.

Author contribution

A.O. Bessatian: conceptualization, systematization of clinical and digital data, analysis, and manuscript writing.

References

• 1. Kafedzhieva A., Vlahova A., Chuchulska B. (2025) Digital technologies in implantology: a narrative review. Bioengineering (Basel), 12(9): 927. DOI: 10.3390/bioengineering12090927.
• 2. Ntovas P., Sirirattanagool P., Asavanamuang P. et al. (2025) Rehabilitation of a terminal dentition patient using an immediate loading protocol with a virtual patient-assisted workflow: a case report. J. Esthet. Restor. Dent., 37(7): 1615–1631. DOI: 10.1111/jerd.13451.
• 3. Chen S., Zeng P., Chen Y. et al. (2026) Construction and validation of a patient-centered data governance platform for multimodal dental implant data management. Clin. Oral Investig., 30(5): 193. DOI: 10.1007/s00784-026-06892-x.
• 4. Zaman G., Khan R.S., Spacey A. et al. (2026) The use of artificial intelligence in planning dental implant procedures: a systematic review. Dent. J. (Basel), 14(5): 248. DOI: 10.3390/dj14050248.
• 5. Mun S.B., Yoo S.R., Kim Y.J. et al. (2025) AI-driven prediction of dental implant numbers to be placed for patient-specific treatment planning. Int. Dent. J., 75(6): 103896. DOI: 10.1016/j.identj.2025.103896.
• 6. Che S.A., Yang B.E., Park S.Y. et al. (2025) Clinical evaluation of AI-based three-dimensional dental implant planning: a multicenter study. J. Dent., 162: 106066. DOI: 10.1016/j.jdent.2025.106066.
• 7. Lin C.C., Lin H.W., Lin Y.C., Lin W.C. (2025) Digital workflow for full-arch immediate loading: dynamic navigation and 3D printing fixed dental prostheses. J. Dent. Sci., 20(4): 2536–2537. DOI: 10.1016/j.jds.2025.06.016.
• 8. de Araújo Nobre M., Almeida R., Moura Guedes C. et al. (2025) Digital workflow for interim prosthetic rehabilitation through the All-on-4 concept using 3D printing additive process. J. Clin. Med., 14(23): 8353. DOI: 10.3390/jcm14238353.
• 9. Nandedkar C., Ronad S., Dodamani G. et al. (2025) Digitally guided multidisciplinary rehabilitation of a partially edentulous patient using implant-supported prostheses: a case report. Cureus, 17(6): e86824. DOI: 10.7759/cureus.86824.
• 10. Valenzuela M.V., Marroquín-Soto C., Padilla-Avalos C.A. (2026) An alternative digital workflow for 3D virtual patient construction integrating intraoral, facial, and CBCT data for esthetic and occlusal planning. J. Esthet. Restor. Dent., 38(8): 1477–1484. DOI: 10.1111/jerd.70154.
• 11. Lepidi L., Kim B.C., Giberti L. et al. (2025) The 4D virtual patient: a proof of concept in digital dentistry. J. Prosthet. Dent., 134(5): 1442–1445. DOI: 10.1016/j.prosdent.2024.02.029.
• 12. Lefrançois E., Delanoue V., Morice S. et al. (2025) A digital approach for a complete rehabilitation with fixed and removable prostheses: a technical procedure. Dent. J. (Basel), 13(1): 7. DOI: 10.3390/dj13010007.
• 13. Etxaniz O., Amezua X., Jauregi M., Solaberrieta E. (2025) Improving the accuracy of complete arch implant digital scans by using auxiliary clips for intraoral scan bodies: a dental technique. J. Prosthet. Dent., 133(1): 43–47. DOI: 10.1016/j.prosdent.2024.01.031.
• 14. Qiu A., Xu L., Zhang Y. et al. (2024) Digital-assisted multidisciplinary treatment for complex occlusal rehabilitation: an 18-month follow-up case report. BMC Oral Health, 24: 818. DOI: 10.1186/s12903-024-04574-2.
• 15. Zaww K., Abbas H., Vanegas Sáenz J.R., Hong G. (2026) AI-driven innovations for dental implant treatment planning: a systematic review. J. Prosthodont. Res., 70(2): 192–205. DOI: 10.2186/jpr.JPR_D_24_00338.
• 16. Pozzi A., Arcuri L., Carosi P. et al. (2025) Photogrammetry versus intraoral scanning in complete-arch digital implant impression: a systematic review and meta-analysis. Clin. Implant Dent. Relat. Res., 27(3): e70059. DOI: 10.1111/cid.70059.
• 17. Borthakur P.P., Sahariah J.J., Sarma M. et al. (2025) Advances in 3D printing for dentistry: clinical applications and future perspectives. Explor. Med., 6: 1001374. DOI: 10.37349/emed.2025.1001374.
• 18. Almarshadi R., Hamdi S., Hadi F. et al. (2025) Redefining digital dentistry: multidisciplinary applications of 3D printing for personalized dental care. Cureus, 17(6): e86791. DOI: 10.7759/cureus.86791.
• 19. Adel H.A., Abuljabbar A.A. (2025) The impact of 3D printing in orthodontics and maxillofacial surgery: applications, challenges, and future directions. UMJ, 1–7. DOI: 10.32471/umj.1680-3051.268261.
• 20. Chackartchi T., Romanos G.E., Parkanyi L. et al. (2022) Reducing errors in guided implant surgery to optimize treatment outcomes. Periodontol. 2000, 88(1): 64–72. DOI: 10.1111/prd.12411.
• 21. Flügge T., Derksen W., te Poel J. et al. (2017) Registration of cone beam computed tomography data and intraoral surface scans — a prerequisite for guided implant surgery with CAD/CAM drilling guides. Clin. Oral Implants Res., 28(9): 1113–1118. DOI: 10.1111/clr.12925.
• 22. Tahmaseb A., Wu V., Wismeijer D. et al. (2018) The accuracy of static computer-aided implant surgery: a systematic review and meta-analysis. Clin. Oral Implants Res., 29(Suppl. 16): 416–435. DOI: 10.1111/clr.13346.
• 23. Khaohoen A., Powcharoen W., Sornsuwan T. et al. (2024) Accuracy of implant placement with computer-aided static, dynamic, and robot-assisted surgery: a systematic review and meta-analysis of clinical trials. BMC Oral Health, 24: 359. DOI: 10.1186/s12903-024-04033-y.
• 24. Riley D.S., Barber M.S., Kienle G.S. et al. (2017) CARE guidelines for case reports: explanation and elaboration document. J. Clin. Epidemiol., 89: 218–235. DOI: 10.1016/j.jclinepi.2017.04.026.

Надійшла до редакції/Received: 21.05.2026
Прийнято до друку/Accepted: 25.05.2026