Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2025 Feb 15;36(6):725–735. doi: 10.1111/clr.14420

Accuracy of a Novel Robot‐Assisted System and Dynamic Navigation System for Dental Implant Placement: A Clinical Retrospective Study

Jianping Chen 1, Yude Ding 1, Ruijue Cao 2, Yuchen Zheng 1, Liheng Shen 1, Linhong Wang 1,✉, Fan Yang 1,✉
PMCID: PMC12146509  PMID: 39953801

ABSTRACT

Aim

This clinical retrospective study aimed to compare the deviations in single‐tooth implant placement using novel semi‐autonomous robotic‐assisted surgery system (sa‐RASS) and dynamic navigation system (DNS) methods.

Materials and Methods

A retrospective analysis of medical records from January to December 2023 was conducted to compare the implantation accuracy of the sa‐RASS and DNS in partially edentulous patients using cone‐beam computed tomography. Platform, apex, and angular deviations were measured and analyzed. The Kolmogorov–Smirnov test was used to check the data distribution, along with t‐tests or Mann–Whitney U‐tests, where appropriate.

Results

Fifty‐seven patients (57 implants) were analyzed: 29 (29 implants) in the sa‐RASS group and 28 (28 implants) in the DNS group. The comparison of platform, apex, and angular deviation between the sa‐RASS group and DNS group were 0.91 ± 0.46 mm vs. 1.26 ± 0.51 mm (p < 0.001), 1.06 ± 0.60 mm vs. 1.51 ± 0.56 mm (p < 0.001), and 3.07° ± 1.66° vs. 3.71° ± 1.64° (p > 0.05), respectively. In addition, there was no significant difference in the accuracy of different implant regions (premolar, molar, maxilla, and mandible) or implant length (p > 0.05).

Conclusions

In the present study, sa‐RASS implant placement showed better positional accuracy than DNS implant placement in platform and apex deviation, although these improvements in accuracy may have limited clinical relevance, suggesting that the sa‐RASS might be advantageous in dental implant surgery.

Trial Registration: ClinicalTrials.gov identifier: ChiCTR2400085089

Keywords: accuracy, dynamic navigation system, implant dentistry, semi‐autonomous robot

1. Introduction

The ideal three‐dimensional (3D) position is required for long‐term stability of the peri‐implant hard and soft tissues, as well as a prerequisite for optimal functional mechanics and aesthetics (Buser et al. 2017; Herrera et al. 2023; Kopp et al. 2003). However, traditional freehand implant surgery mainly relies on the clinician's clinical experience, making accurate implant placement challenging. Furthermore, traditional freehand implant surgery increases the risk of damage to important adjacent anatomical structures. The incidence of complications resulting from positioning failure has been reported to be more than 10%, even for experienced surgeons, when placing implants with free hands (Naeini et al. 2020; Chee and Jivraj 2007). Presurgical planning combined with use of a surgical guide during the placement of dental implants is important for achieving a precise implant position. The manufacture of surgical guides based on computed tomography, computer‐assisted surgery, static computer‐assisted implantation surgery (sCAIS), and dynamic computer‐assisted implantation surgery (dCAIS) have been applied to osteotomy preparation and implant placement (Jorba‐García et al. 2021). Research has indicated that sCAIS and dCAI enhance implant accuracy significantly when compared to freehand implantation (Kaewsiri et al. 2019; Kivovics et al. 2022). The mean platform, apex, and angular deviations in sCAIS and dCAIS were reported to be significantly lower than those of freehand (1.20 mm or 1.16 mm vs. 1.78 mm, 1.40 mm or 1.29 mm vs. 2.27 mm, and 3.50° or 2.97° vs. 6.50°, respectively) (Block et al. 2017b; Tahmaseb et al. 2018).

However, sCAIS has some limitations. In clinical practice, implant sites with narrow gaps and limited posterior openings can be challenging to manipulate using sCAIS. There are also issues with the inability to visualize the operative area, the difficulty of water cooling, and the additional cost and long lead time required for the time‐consuming fabrication (D'haese et al. 2017; Sigcho López et al. 2019). Although dCAIS has advantages such as being intuitive, controllable during usage, and verifiable at any time, it also has many drawbacks, including system and operational failures. Dynamic navigation implant surgery remains a freehand operation, influenced by hand tremors and sensations, and has a steep learning curve (Block et al. 2017b; Sun et al. 2019). Therefore, a method aimed at improving the stability and reducing the vibration of the drill system is needed to reduce reliance on highly skilled clinicians. Robotic arms can address some of these shortcomings.

A robot‐assisted system incorporating haptic constraints and real‐time feedback is a highly accurate and stable solution for dental implant placement. Dental implant robots can be categorized as passive (master–slave, collaborative) and autonomous. Master–slave involves a person remotely controlling the robotic arm, collaborative involves a person directly manipulating the robotic arm and collaborating to complete the implantation surgery, and autonomous involves the robotic arm autonomously and independently completing the implantation surgery (Li et al. 2020). A case study conducted by Mozer et al. using a master–slave implant robotic system, Yomi (Neocis Inc., Miami, FL, USA), demonstrated that the platform deviation of the robotic system was approximately 0.79 mm, the apex deviation was 0.87 mm, and the angular deviation was 1° (Mozer 2020). Jia et al. evaluated the accuracy of an autonomous dental implant robotic (ADIR) system for the restoration of implants in a retrospective study, in which 19 participants received the ADIR system and reported that the platform, apex, and angular deviations were 0.43 ± 0.18 mm, 0.56 ± 0.18 mm, and 1.48° ± 0.18°, respectively (Jia et al. 2023).

Although both dynamic navigation systems and robot‐assisted implant systems have demonstrated acceptable implantation accuracy, our previous work on model analyses showed higher accuracy for robot‐assisted implant placement than for dynamic navigation systems (Chen et al. 2023b). However, there is no high‐level clinical evidence comparing robot‐assisted and dynamic navigation implant placement. Therefore, this study aimed to compare the accuracy of a novel semi‐autonomous robotic‐assisted surgery system (sa‐RASS) with that of a dynamic navigation system (DNS) by evaluating the deviations between preoperative planning and postoperative placement.

2. Materials and Methods

2.1. Study Design and Ethical Approval

The medical records of partially edentulous patients who underwent implant surgery using the sa‐RASS system (Hangzhou Jianjia Medical Technology Co. Ltd.) or the DNS system (Suzhou Digital Medical Co. Ltd.) at the Stomatology Center of Zhejiang Provincial People's Hospital from January to December 2023 were retrospectively analyzed to collect demographic information and detailed medical records. The present study complied with the Declaration of Helsinki was registered in the Chinese Clinical Trial Registry and was approved by the local ethics committee of Zhejiang Provincial People's Hospital, China (No. 2024118). All patients provided informed consent prior to the surgery. The study complied with the privacy rights of the participants. This study complied with the STROBE checklist.

The inclusion criteria were as follows: (1) Implants placed using the sa‐RASS or DNS system, (2) partially edentulous patients, and (3) accessible and comprehensive record information. The exclusion criteria were as follows: (1) Implant implantation was not performed according to the preoperative plan and (2) the postoperative cone‐beam computed tomography (CBCT) images were not good enough to accurately identify the implant site (Figure S1).

2.2. Preoperative Procedure of Sa‐RASS and DNS

The sa‐RASS comprises visual equipment, a robotic arm and terminal actuator, a patient‐tracking reflective device, and a control system (Figure 1). A U‐tube containing silicone impression material (Silagum, DMG, Hamburg, Germany) was worn at the missing tooth location, and CBCT (Planmeca ProMax, Planmeca Oy, Helsinki, Finland) was performed and image data in the Digital Imaging and Communications in Medicine (DICOM) format were obtained. All scans were performed at 80 kV and 6.0 mA for 15 s (voxel size: 0.15 mm; grayscale: 15 bits; focal spot: 0.5 mm; and field of view: 12 × 9 cm). A dental scanner (TRIOS‐3 basic intraoral scanner, 3Shape A/S) was used to obtain intraoral information in a standard inlay language (STL) format. An experienced operator entered the DICOM files of the CBCT data into the sa‐RASS system software (Cycad DHC‐DI, version V1, Hangzhou Jianjia Medical Technology Co. Ltd. China) to design the implant planning position according to the prosthetic‐driven principle.

FIGURE 1.

FIGURE 1

Open in a new tab

Architecture of semi‐autonomous robotic‐assisted surgery system.

The DNS (DHC‐D12, Suzhou Digital‐health Care Co. Ltd., China) included an infrared optical tracker, an implant handpiece with a biplane positioner, a reference plate, a U‐tube, a drilling needle, and navigation software (Figure 2). Similar to the sa‐RASS, the DNS steps were guided by the prosthetic‐driven principle. Initially, a reference marker was securely attached to the teeth near the surgical area using an acrylic resin (Silagum, DMG, Hamburg, Germany). Prior to surgery, CBCT was performed to obtain preoperative images. The DICOM files of these scans were imported into the DNS design software. An experienced operator used the design software to plan the 3D positions of the implants and the sequence of the operation. The preoperative planning procedure was completed before surgery. The experimental procedures for the two systems are shown in Figure 3.

FIGURE 2.

FIGURE 2

Open in a new tab

Architecture of dynamic navigation system.

FIGURE 3.

FIGURE 3

Open in a new tab

Experimental operation procedure for sa‐RASS and DNS.

2.3. Surgical Procedure

2.3.1. Sa‐RASS

The end arrays and handpiece were calibrated to ensure that the position of the robotic arm could be captured and confirmed during surgery. After standard disinfection, local anesthesia was induced in patients using articaine in the mucosa of the missing tooth area (4% articaine hydrochloride, adrenaline, 1:100 000). The U‐tube was repositioned in the missing tooth area, and the position marker was fixed using self‐curing acrylic resin (Protemp, 3 M ESPE, Neuss, Germany) on the area of the residual teeth on the opposite side of the same jaw. After U‐tube registration and accuracy verification were completed, a full‐thickness gingival flap was performed at the target site of the implant. In this study, the implant surgeon (D.Y.D.) was implantology certified and had over 8 years of implant surgery expertise. The surgeon was proficient in both robot operation and dynamic navigation after training. The surgeon pressed the teach button to freely drag the robotic arm to an appropriate position close to the oral region, ensuring that the distance did not exceed 10 mm. Subsequently, the surgeon released the teach button, allowing the robotic arm to automatically position itself at the target location of the missing tooth position. Once automatic positioning was complete, the surgeon pressed and held the alignment button to perform alignment movements along the planned axis of the implant. After completing each drilling level, the surgeon released the alignment button, pressed the teach button, removed the robotic arm from the mouth, and replaced the drill. No bone augmentation was performed. Finally, the surgeon inserted the implant (Nobel Parallel CC, Nobel Biocare, Zurich, Switzerland) using a robotic arm with a torque of 35 Ncm.

Throughout the drilling process, the control panel on the right displayed real‐time information such as the drilling lateral deviation, angle deviation, and drilling depth. (Figure 4).

FIGURE 4.

FIGURE 4

Open in a new tab

Workflow of sa‐RASS method. (a) Wearing of the U‐tube; (b) implant position planning; (c) intraoral alignment and validation of the U‐tube; (d) U‐tube registration and validation operation panel; (e) sa‐RASS implant cavity preparation and implant placement; (f) real‐time display of the sa‐RASS screen; (g) fusion and calibration of the preoperative and postoperative images using the 3D Slicer software; and (h) measurement of preoperative and postoperative deviations.

2.3.2. DNS

The infrared optical tracker was matched to the handpiece reference and registration plates. After standard disinfection, local anesthesia was induced in patients using articaine in the mucosa of the missing tooth area (4% articaine hydrochloride; adrenaline, 1:100 000). According to the position of the jaw and the target surgical area, the infrared optical tracker was placed at a distance of approximately 1.25–1.50 m and 45°–60° from the front of the patient's head in order to capture the position of the implant handpiece and the patient's jaw. After calibration, the fixed reference plate was attached with self‐curing acrylic resin (Protemp, 3 M ESPE, Neuss, Germany) to the residual teeth area on the opposite side of the same jawbone. The U‐tube was then repositioned in the missing tooth area, and registration was performed. The alignment error was controlled to within 0.1 mm. Additionally, the accuracy of the registration could be verified for the adjacent teeth simultaneously. A full‐thickness gingival flap was performed at the target site of the implant. Guided by the DNS system, the surgeon (D.Y.D.) performed drilling and implant insertion. No bone augmentation was performed. The screen of the DNS showed the real‐time situation of the implant surgery area and guided the surgeon to follow the preoperative design, ensuring the precise transfer of the preoperative plan to the intraoperative stage (Figure 5).

FIGURE 5.

FIGURE 5

Open in a new tab

Workflow of DNS method. (a) Wearing of the U‐tube; (b) implant location planning; (c) intraoral reference plate fixation and U‐tube registration; (d) screenshot of the U‐tube registration; (e) DNS implant cavity preparation and implant placement; (f) real‐time display of the DNS screen; (g) fusion and calibration of the preoperative and postoperative images using the 3D Slicer software; and (h) measurement of preoperative and postoperative deviations.

2.4. Postoperative Treatment

All postoperative incisions with missing teeth were tightly sutured using 4–0 Prolene. All patients were administered antibiotics, including cefuroxime axetil (250 mg, twice a day, 3–5 days; Shenzhen Zhijun Pharmaceutical Co. Ltd.) and metronidazole (250 mg, three times a day, 3–5 days; Sichuan Kelun Pharmaceutical Group Co. Ltd.). A 0.1% cetylpyridinium chloride gargle (Hangzhou Minsheng Pharmaceutical Co. Ltd.) was suggested 1 week after surgery. The sutures were removed 1 week after surgery.

2.5. Accuracy Measurement

CBCT was performed immediately after surgery using the same parameters as the CBCT for preoperative planning. Preoperatively planned implant 3D images and postoperatively placed CBCT images were uploaded to 3D Slicer software (version 4.13; Harvard University, Boston, USA), as described by Talmazov et al. (2020). First, the preoperative CBCT scans containing the coordinates of the planned implant position and the postoperative CBCT scans were imported into the 3D Slicer software. Then, the preoperative and postoperative 3D reconstructed models were superimposed using three alignment points at the same position. Precise alignment was obtained by the point cloud acquisition algorithms of 3D Slicer and manually adjusted if necessary. After the automatic recognition of the actual implant position, the coordinates of the planned and actual implants were compared (Figures 4g,h and 5g,h). Implant position accuracy measurements, including platform deviation (mm), apex deviation (mm), and angular deviation (°), were performed by a trained medical professional. The professional was blinded to the surgical information and grouping. For the platform and apex points, the global, horizontal, and vertical deviations were analyzed separately (Figure 6). All the measurements were performed by a trained operator. Duplicate registration was carried out at intervals of 24 h by the same operator to evaluate intra‐operator dependability. The deviation was computed using the mean number of observations.

FIGURE 6.

FIGURE 6

Open in a new tab

Measurement parameters of the pre‐ and postoperative implant deviations.

2.6. Statistical Analysis

The sample size was calculated using G*Power v.3.1.3 (Heinrich‐Heine Universität, Düsseldorf, Germany). Based on the implant platform, apex, and angle deviation values of robotic‐assisted surgery and dynamic navigation systems reported in previous studies (0.68 ± 0.36 mm vs. 1.25 ± 0.54 mm, 0.69 ± 0.36 mm vs. 1.39 ± 0.52 mm, and 1.37° ± 0.92° vs. 4.09° ± 1.79°, respectively) (Zhang et al. 2023), the minimum required sample size was 11, 17, and 9 implants according to platform, apex, and angle deviation, respectively. An alpha value of 0.05 and a statistical power of 90% were established. The final sample size was determined to be at least 17 patients in each group for a total of 34 patients. Statistical analyses were performed using SPSS software (version 25.0; IBM Corp., Armonk, NY, USA) (IBM SPSS Statistics Inc., V25; IBM Corporation). The Shapiro–Wilk test was used to assess the normal distribution of the data. Descriptive analyses were performed using the mean and standard deviation for normally distributed data. For non‐normally distributed data, descriptive analyses were performed using the median and interquartile range. The student's t‐test or Mann–Whitney U‐test was applied to compare the deviations of the sa‐RASS and DNS, as appropriate. A linear regression model was established to analyze variations in deviation. The statistical significance level for all analyses was set at p < 0.05.

3. Results

Data from 57 partially edentulous patients (57 implants) were analyzed: 29 patients (29 implants) in the sa‐RASS group and 28 patients (28 implants) in the DNS group. No adverse surgical events or postoperative complications were observed. The descriptive data of the patients are presented in Table 1.

TABLE 1.

Demographic and clinical characteristics of the included patients.

sa‐RASS DNS p
Age
Median (range) 32 (22–65) 51 (22–68)
Mean ± SD 44.00 ± 8.00 47.75 ± 5.76 p < 0.01**
Gender
Male 9 10
Female 20 18
Total number of implants 29 28
Implant position
Maxillary
Premolar 2 2
Molar 5 2
Mandible
Premolar 4 3
Molar 18 21

Implant diameter

3.75/4.3/5.0 mm

 1/23/5 2/19/7

Implant length

8.5/10/11.5 mm

 10/18/1 9/16/3
Open in a new tab

Abbreviations: sa‐RASS, semi‐autonomous robotic‐assisted surgery system; DNS, dynamic navigation system; SD, standard deviation.

**

p < 0.01.

One operator repeated the evaluations and calculations, and the intraclass correlation was greater than 0.95 for all measurements. The results of the linear regression model showed that different implant regions (premolar, molar, maxilla, and mandible) and implant length did not significantly influence the accuracy of implant placement (p > 0.05) (Table 2).

TABLE 2.

Linear regression analysis with different variables.

Linear regression model Platform global deviation Apex global deviation Angular deviation Platform vertical deviation Platform horizontal deviation Apex vertical deviation Apex horizontal deviation
R 2 0.145 0.176 0.106 0.056 0.111 0.075 0.157
p p p p p p p
Group 0.0087** 0.0055** 0.1557 0.0612 0.0474* 0.1204 0.0169*
Premolar/Molar 0.902 0.249 0.341 0.992 0.954 0.532 0.197
Maxilla/mandible 0.161 0.206 0.634 0.062 0.718 0.080 0.547
Implant length (8.5/10/11.5 mm) > 0.99 > 0.99 0.177 > 0.99 0.999 0.998 0.992
Open in a new tab
*

p < 0.05.

**

p < 0.01.

The platform global, apex global, and angular deviation of the sa‐RASS group and DNS group were 0.91 ± 0.46 mm vs. 1.26 ± 0.51 mm (p < 0.01), 1.06 ± 0.60 vs. 1.51 ± 0.56 mm (p < 0.01), and 3.07° ± 1.66° vs. 3.71° ± 1.64° (p > 0.05), respectively. Statistically significant differences were found between the two groups in the platform global, apex global, and horizontal deviations; however, there were no statistical differences in the angular and vertical deviations (Table 3, Figures 7 and 8).

TABLE 3.

Implant accuracy.

Platform deviation (mm) Apex deviation (mm) Angular deviation (°)
Group Global Vertical Horizontal Global Vertical Horizontal
sa‐RASS Median 0.96 0.51 0.73 0.94 0.46 0.76 2.34
25th percentile 0.53 0.20 0.32 0.67 0.16 0.56 1.81
75th percentile 1.20 0.79 1.00 1.33 0.76 1.20 4.64
Range 1.94 1.44 1.39 3.11 2.38 2.24 4.45
Mean ± SD 0.91 ± 0.46 0.51 ± 0.37 0.70 ± 0.39 1.06 ± 0.60 0.52 ± 0.49 0.88 ± 0.47 3.07 ± 1.66
DNS Median 1.19 0.72 0.83 1.51 0.73 1.22 3.49
25th percentile 0.89 0.39 0.63 1.08 0.42 0.74 2.79
75th percentile 1.49 0.92 1.15 1.90 0.93 1.73 4.43
Range 2.20 1.39 2.40 2.28 1.37 2.40 6.40
Mean ± SD 1.26 ± 0.51 0.69 ± 0.35 0.96 ± 0.56 1.51 ± 0.56 0.70 ± 0.34 1.26 ± 0.64 3.71 ± 1.64
MD 0.36 ± 0.65 0.20 ± 0.49 0.26 ± 0.65 0.45 ± 0.72 0.19 ± 0.58 0.37 ± 0.73 0.60 ± 2.35
p 0.0087** 0.0612 0.0474* 0.0055** 0.1204 0.0169* 0.1557
Open in a new tab

Abbreviations: DNS, dynamic navigation system; MD, mean difference; sa‐RASS, semi‐autonomous robotic‐assisted surgery system; SD, standard deviation.

*

p < 0.05.

**

p < 0.01.

FIGURE 7.

FIGURE 7

Open in a new tab

Boxplots showing distribution of angular deviation between sa‐RASS and DNS; sa‐RASS, semi‐autonomous robotic‐assisted surgery system; DNS, dynamic navigation system.

FIGURE 8.

FIGURE 8

Open in a new tab

Boxplots showing distribution of linear deviations between sa‐RASS and DNS groups. AG, apex global; AV, apex vertical; AH, apex horizontal; DNS, dynamic navigation system; sa‐RASS, semi‐autonomous robotic‐assisted surgery system; PG, platform global; PV, platform vertical; PH, platform horizontal.

4. Discussion

Our results showed that the sa‐RASS technology had mean global platform, global apex, and angular deviations of 0.91 mm, 1.06 mm, and 3.07°, respectively. Additionally, no intraoperative complications were reported. Some systematic reviews and meta‐analyses on the accuracy of surgical guide templates showed that the mean platform and apex deviations of computer‐aided implant surgery were less than 1.22 mm and 1.45 mm, respectively, and the angular deviation was < 4.06° (Bover‐Ramos et al. 2018; Tahmaseb et al. 2018). In another clinical study, the deviation of implants placed by freehand was reported as high as 1.70 mm at the platform, 2.51 mm at the apex, and 10.04° at the angulation (Aydemir and Arısan 2019). Compared with the reported implantation deviation of conventional techniques, that of sa‐RASS technology make it a potential alternative to conventional s‐CAIS and freehand implants, offering more precise and predictable approaches (Yang et al. 2024). Although the accuracy improvement of the sa‐RASS technology in this study was minimal compared to that of DNS, in some cases with anatomical limitations in clinical practice, the robotic system improved the accuracy of implant placement and prevented damage to key anatomical structures, such as the inferior alveolar nerve and maxillary sinus. Additionally, DNS does not provide a method to improve stabilization and decrease the vibration of the drill system and thus relies on the clinical expertise of a highly skilled clinician, whereas the robotic arm may address these shortcomings. Finally, the robotic system runs under the control of human‐robot collaboration. Surgeons can stop the drilling procedures when needed, thereby greatly improving the safety of the surgery. For novice surgeons, robot‐assisted surgery is an excellent option (Yang et al. 2024b). According to another in vivo study of the ADIR system, the average platform, apex, and angular deviations were 0.43 ± 0.18 mm, 0.56 ± 0.18 mm, and 1.48° ± 0.18°, respectively, which were more accurate than those of sa‐RASS. This is because ADIR integrates visual sensing, force sensing, 3D visualization, and micromodular robotics to achieve autonomous robot positioning, autonomous cavity import and export, autonomous preparation of implantation sites, autonomous implant placement, and other operations. This minimizes the implantation deviation caused by the surgeon's lack of surgical experience (Chen et al. 2023a; Zhang et al. 2023). The arms of the semiautonomous robot used in this study were under the control of the surgeons. This design allows surgeons to respond quickly to sudden and noticeable changes in the patient's head position during surgery, further ensuring emergency safety. These deviations in sa‐RASS can be attributed to the following factors. First, the surgeon is more focused on the surgical site and does not need to visually alternate between the display screen and surgical site, thereby minimizing distraction (Xu et al. 2023). Second, sa‐RASS improves stability and joint flexibility, allowing for real‐time position capture through optical tracking and localization, and automatic calibration of slight movements of the patient's head to ensure accurate surgical procedures (Chen et al. 2023a). Importantly, the stability of the end robotic arm avoids operator error during implantation and drilling due to operator fatigue, hand tremors, blind spots, and poor body position and reduces the complexity of the treatment. Hand tremors and inaccurate perceptions have been shown to lead to lateral deviations of 0.25 mm and angular deviations of 0.5° (Parra‐Tresserra et al. 2021; Ruppin et al. 2008). Our results showed that, in the DNS group, platform global deviation was 1.26 ± 0.51 mm, apex global deviation was 1.51 ± 0.56 mm, and angular deviation was 3.71° ± 1.64°, which is in accordance with recent systematic reviews showing that the average platform, apex, and angular deviations of DNS are approximately 0.5–1.5 mm, 0.5–2.0 mm, and 2°–7°, respectively (Pellegrino et al. 2021; Yu et al. 2023). The common reasons for the deviations caused by dynamic navigation techniques are the weight and volume of the mobile positioner affecting operational stability and increasing fatigue for the operating surgeon during drilling and implant placement (Block et al. 2017a; Chen et al. 2023b), instability of the fixation device and any minor looseness or movement between the relevant components due to incorrect calibration (Parra‐Tresserra et al. 2021), and the accuracy and quality of scan data due to different CBCT slice thicknesses and voxels (Zhao et al. 2014).

In addition, we found that the DNS group had a higher deviation at the apex than at the platform in both the global and horizontal dimensions, which was not significant in the sa‐RASS group. This may be because the angular deviation was higher in the DNS group than in the sa‐RASS group based on the deviation at the platform, and the deviation of the apex position increased the angular factor. In both systems, the vertical and horizontal deviations exhibited opposite results for the platform and apex regions. The vertical deviation of the apex was smaller than that of the platform because the operator usually monitors the vertical deviation of the apex of the implant as a reference for the implant position. The vertical position of the platform is important in the final restoration, suggesting that special attention should be paid to the depth of the platform, especially in the esthetic restoration of the anterior teeth. The horizontal deviation was smaller at the platform than at the apex because the position of the drill at the platform was more visible, controllable, and easier to change than that at the apex.

The influence of the trial group and implant region (premolar, molar, maxilla, mandible, and implant length) on the main outcome variables was analyzed using multiple linear regression. The results showed a statistically significant effect in the clinical trial group (p < 0.05), whereas the differences among the various implant regions were not statistically significant (p > 0.05). These findings suggest consistent implant precision with appropriate implant length. However, previous studies have indicated that implant length may influence apex deviation, possibly because of the fixed platform side of the guide and the inability to fix the apex side (Bolding and Reebye 2022). Additionally, deviations toward areas of bone with lower resistance during osteotomy preparation with drills may also contribute to deviations from the originally planned implant position (Jia et al. 2023). Although both systems achieve high accuracy, unavoidable errors can occur during use. Both systems register reference materials for tooth retention, which may shift slightly during use, and the movements of the marks and inaccuracy of registration also result in deviations (Ma et al. 2022). In addition, the accuracy of the sa‐RASS method in this study may have been affected by the limited technology and the oral condition of the patient, and the robotic arm was still not sufficiently flexible. It operates along a preset path and cannot change the inclination angle when drilling to overcome obstacles, thereby increasing difficulty when used in the posterior region with limited interocclusal space (Shi et al. 2024). This study has some limitations. First, because CBCT scanning was used to evaluate the accuracy deviation, the main limitation was the radiographic artifacts that were more often observed in the apex and middle parts of the mandible and in the anterior region of the jaws (Terrabuio et al. 2021). In future investigations, there should be a shift from the use of postoperative CBCT to less invasive methods, such as digital intraoral scans, to avoid additional radiation. Second, this study only compared the accuracy of implant placement in partially edentulous patients. Therefore, a comparative analysis of the reliability and feasibility of the sa‐RASS in completely edentulous patients should be conducted. Third, although robotic‐assisted implant surgery has demonstrated excellent accuracy, its clinical practicality is constrained by factors such as time efficiency. Preoperative preparation for robotic surgery involves multiple steps including implant design, calibration, registration, and verification, which significantly prolong the overall preparation process. Regarding operative time, previous studies reported that the autonomous robotic implant surgery group required considerably longer drilling and implant placement durations than the dynamic navigation implant surgery group (10.6 ± 3.8 vs. 8.3 ± 3.4 min, p < 0.01), while preparation time showed no statistical difference between the groups (7.2 ± 3.3 vs. 6.2 ± 2.7 min, p > 0.05), and was significantly longer than that of freehand implant surgery (3.3 ± 0.6 min) (p < 0.001) (Neuschitzer et al. 2025; Yu et al. 2024). Finally, this was a retrospective study with a small sample size. Future research should use randomized controlled trials to obtain more scientific findings.

5. Conclusions

Within the limitations of this clinical study, semiautonomous robotic‐assisted dental implant placement demonstrated greater implant accuracy than dynamic navigation, particularly in terms of platform and apex deviations. Although some of these improvements in accuracy were statistically significant, their clinical relevance may have been limited by the small magnitude of the observed differences. Further development and additional studies are required to promote the widespread application of robot‐assisted dental implant surgery in clinical practice.

Author Contributions

Jianping Chen: conceptualization, validation, methodology, formal analysis, writing – original draft, investigation. Yude Ding: conceptualization, investigation, formal analysis, methodology, writing – original draft, visualization. Ruijue Cao: validation, investigation, methodology, data curation, writing – original draft, software. Yuchen Zheng: writing – review and editing, data curation, investigation, software, funding acquisition. Liheng Shen: software, writing – review and editing, data curation, visualization. Linhong Wang: writing – review and editing, funding acquisition, visualization, project administration, resources, supervision. Fan Yang: funding acquisition, writing – review and editing, visualization, project administration, supervision, resources, conceptualization.

Ethics Statement

The study was registered in the Chinese Clinical Trial Registry and approved by the local ethics committee of Zhejiang Provincial People's Hospital, China. (No. 2024118).

Consent

Investigators obtained informed consent before enrolling participants in clinical trials.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1. Workflow of patients’ enrolment.

CLR-36-725-s001.docx (97.4KB, docx)

Acknowledgements

The authors would like to acknowledge the excellent technical assistance of Mijia Dai and Chao Cheng from Hangzhou Jianjia Medical Technology Co. Ltd., as well as the collaboration of other team members.

Funding: This study was supported by the Zhejiang Provincial Key R&D Program (Leading Goose Projects) (2024C03094) and Major Scientific and Technological Project of Zhejiang Province (WKJ‐ZJ‐2328).

Jianping Chen, Yude Ding, Ruijue Cao contributed equally to this work.

Contributor Information

Linhong Wang, Email: wanglinhong@hmc.edu.cn.

Fan Yang, Email: yangfan@hmc.edu.cn.

Data Availability Statement

The datasets used and/or analyzed in the current study are available from the corresponding author upon reasonable request.

References

  1. Aydemir, C. A. , and Arısan V.. 2019. “Accuracy of Dental Implant Placement via Dynamic Navigation or the Freehand Method: A Split‐Mouth Randomized Controlled Clinical Trial.” Clinical Oral Implants Research 31, no. 3: 255–263. 10.1111/clr.13563. [DOI] [PubMed] [Google Scholar]
  2. Block, M. , Emery R., Lank K., and Ryan J.. 2017a. “Implant Placement Accuracy Using Dynamic Navigation.” International Journal of Oral & Maxillofacial Implants 32, no. 1: 92–99. 10.11607/jomi.5004. [DOI] [PubMed] [Google Scholar]
  3. Block, M. S. , Emery R. W., Cullum D. R., and Sheikh A.. 2017b. “Implant Placement Is More Accurate Using Dynamic Navigation.” Journal of Oral and Maxillofacial Surgery 75, no. 7: 1377–1386. 10.1016/j.joms.2017.02.026. [DOI] [PubMed] [Google Scholar]
  4. Bolding, S. L. , and Reebye U. N.. 2022. “Accuracy of Haptic Robotic Guidance of Dental Implant Surgery for Completely Edentulous Arches.” Journal of Prosthetic Dentistry 128, no. 4: 639–647. 10.1016/j.prosdent.2020.12.048. [DOI] [PubMed] [Google Scholar]
  5. Bover‐Ramos, F. , Viña‐Almunia J., Cervera‐Ballester J., Peñarrocha‐Diago M., and García‐Mira B.. 2018. “Accuracy of Implant Placement With Computer‐Guided Surgery: A Systematic Review and Meta‐Analysis Comparing Cadaver, Clinical, and in Vitro Studies.” International Journal of Oral & Maxillofacial Implants 33, no. 1: 101–115. 10.11607/jomi.5556. [DOI] [PubMed] [Google Scholar]
  6. Buser, D. , Sennerby L., and Bruyn H. D.. 2017. “Modern Implant Dentistry Based on Osseointegration 50 Years of Progress, Current Trends and Open Questions.” Periodontology 2000 73, no. 1: 7–21. 10.1111/prd.12185. [DOI] [PubMed] [Google Scholar]
  7. Chee, W. , and Jivraj S.. 2007. “Failures in implant dentistry.” British Dental Journal 202, no. 3: 123–129. 10.1038/bdj.2007.74. [DOI] [PubMed] [Google Scholar]
  8. Chen, J. , Bai X., Ding Y., et al. 2023b. “Comparison the Accuracy of a Novel Implant Robot Surgery and Dynamic Navigation System in Dental Implant Surgery: An In Vitro Pilot Study.” BMC Oral Health 23, no. 1: 179. 10.1186/s12903-023-02873-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Chen, W. , Al‐Taezi K. A., Chu C. H., et al. 2023a. “Accuracy of Dental Implant Placement With a Robotic System in Partially Edentulous Patients: A Prospective, Single‐Arm Clinical Trial.” Clinical Oral Implants Research 34, no. 7: 707–718. 10.1111/clr.14083. [DOI] [PubMed] [Google Scholar]
  10. D'haese, J. , Ackhurst J., Wismeijer D., Bruyn H. D., and Tahmaseb A.. 2017. “Current State of the Art of Computer‐Guided Implant Surgery.” Periodontology 2000 73, no. 1: 121–133. 10.1111/prd.12175. [DOI] [PubMed] [Google Scholar]
  11. Herrera, D. , Berglundh T., Schwarz F., et al. 2023. “Prevention and Treatment of Peri‐Implant Diseases—The EFP S3 Level Clinical Practice Guideline.” Journal of Clinical Periodontology 50, no. S26: 4–76. 10.1111/jcpe.13823. [DOI] [PubMed] [Google Scholar]
  12. Jia, S. , Wang G., Zhao Y., and Wang X.. 2023. “Accuracy of an Autonomous Dental Implant Robotic System Versus Static Guide‐Assisted Implant Surgery: A Retrospective Clinical Study.” Journal of Prosthetic Dentistry. 10.1016/j.prosdent.2023.04.027. [DOI] [PubMed] [Google Scholar]
  13. Jorba‐García, A. , González‐Barnadas A., Camps‐Font O., Figueiredo R., and Valmaseda‐Castellón E.. 2021. “Accuracy Assessment of Dynamic Computer–Aided Implant Placement: A Systematic Review and Meta‐Analysis.” Clinical Oral Investigations 25, no. 5: 2479–2494. 10.1007/s00784-021-03833-8. [DOI] [PubMed] [Google Scholar]
  14. Kaewsiri, D. , Panmekiate S., Subbalekha K., Mattheos N., and Pimkhaokham A.. 2019. “The Accuracy of Static vs. Dynamic Computer‐Assisted Implant Surgery in Single Tooth Space: A Randomized Controlled Trial.” Clinical Oral Implants Research 30, no. 6: 505–514. 10.1111/clr.13435. [DOI] [PubMed] [Google Scholar]
  15. Kivovics, M. , Takács A., Pénzes D., Németh O., and Mijiritsky E.. 2022. “Accuracy of Dental Implant Placement Using Augmented Reality‐Based Navigation, Static Computer Assisted Implant Surgery, and the Free‐Hand Method—An In Vitro Study.” Journal of Dentistry 119: 104070. 10.1016/j.jdent.2022.104070. [DOI] [PubMed] [Google Scholar]
  16. Kopp, K. C. , Koslow A. H., and Abdo O. S.. 2003. “Predictable Implant Placement With a Diagnostic/Surgical Template and Advanced Radiographic Imaging.” Journal of Prosthetic Dentistry 89, no. 6: 611–615. 10.1016/s0022-3913(03)00198-7. [DOI] [PubMed] [Google Scholar]
  17. Li, Y. , Hu J., Tao B., et al. 2020. “Automatic Robot‐World Calibration in an Optical‐Navigated Surgical Robot System and Its Application for Oral Implant Placement.” International Journal of Computer Assisted Radiology and Surgery 15, no. 10: 1685–1692. 10.1007/s11548-020-02232-w. [DOI] [PubMed] [Google Scholar]
  18. Ma, F. , Sun F., Wei T., and Ma Y.. 2022. “Comparison of the Accuracy of Two Different Dynamic Navigation System Registration Methods for Dental Implant Placement: A Retrospective Study.” Clinical Implant Dentistry and Related Research 24, no. 3: 352–360. 10.1111/cid.13090. [DOI] [PubMed] [Google Scholar]
  19. Mozer, P. 2020. “Accuracy and Deviation Analysis of Static and Robotic Guided Implant Surgery: A Case Study.” International Journal of Oral & Maxillofacial Implants 35, no. 5: e86‐e90. 10.11607/jomi.8231. [DOI] [PubMed] [Google Scholar]
  20. Naeini, E. N. , Atashkadeh M., De Bruyn H., and D'Haese J.. 2020. “Narrative Review Regarding the Applicability, Accuracy, and Clinical Outcome of Flapless Implant Surgery With or Without Computer Guidance.” Clinical Implant Dentistry and Related Research 22, no. 4: 454–467. 10.1111/cid.12901. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Neuschitzer, M. , Toledano‐Serrabona J., Jorba‐García A., Bara‐Casaus J. J., Figueiredo R., and Valmaseda‐Castellón E.. 2025. “Comparative Accuracy of dCAIS and Freehand Techniques for Immediate Implant Placement in the Maxillary Aesthetic Zone: An In Vitro Study.” Journal of Dentistry 153: 105472. 10.1016/j.jdent.2024.105472. [DOI] [PubMed] [Google Scholar]
  22. Parra‐Tresserra, A. , Marquès‐Guasch J., Ortega‐Martínez J., Basilio‐Monné J., and Hernández‐Alfaro F.. 2021. “Current State of Dynamic Surgery. A Literature Review.” Medicina Oral, Patología Oral y Cirugía Bucal 26: e576‐e581. 10.4317/medoral.24566. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Pellegrino, G. , Ferri A., Del Fabbro M., Prati C., Gandolfi M., and Marchetti C.. 2021. “Dynamic Navigation in Implant Dentistry: A Systematic Review and Meta‐Analysis.” International Journal of Oral & Maxillofacial Implants 36, no. 5: e121‐e140. 10.11607/jomi.8770. [DOI] [PubMed] [Google Scholar]
  24. Ruppin, J. , Popovic A., Strauss M., Spüntrup E., Steiner A., and Stoll C.. 2008. “Evaluation of the Accuracy of Three Different Computer‐Aided Surgery Systems in Dental Implantology: Optical Tracking vs. Stereolithographic Splint Systems.” Clinical Oral Implants Research 19, no. 7: 709–716. 10.1111/j.1600-0501.2007.01430.x. [DOI] [PubMed] [Google Scholar]
  25. Shi, J.‐Y. , Liu B.‐L., Wu X.‐Y., et al. 2024. “Improved Positional Accuracy of Dental Implant Placement Using a Haptic and Machine‐Vision‐Controlled Collaborative Surgery Robot: A Pilot Randomized Controlled Trial.” Journal of Clinical Periodontology 51, no. 1: 24–32. 10.1111/jcpe.13893. [DOI] [PubMed] [Google Scholar]
  26. Sigcho López, D. A. , García I., Da Silva Salomao G., and Cruz Laganá D.. 2019. “Potential Deviation Factors Affecting Stereolithographic Surgical Guides.” Implant Dentistry 28, no. 1: 68–73. 10.1097/id.0000000000000853. [DOI] [PubMed] [Google Scholar]
  27. Sun, T.‐M. , Lee H.‐E., and Lan T.‐H.. 2019. “The Influence of Dental Experience on a Dental Implant Navigation System.” BMC Oral Health 19, no. 1: 222. 10.1186/s12903-019-0914-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Tahmaseb, A. , Wu V., Wismeijer D., Coucke W., and Evans C.. 2018. “The Accuracy of Static Computer‐Aided Implant Surgery: A Systematic Review and Meta‐Analysis.” Clinical Oral Implants Research 29, no. S16: 416–435. 10.1111/clr.13346. [DOI] [PubMed] [Google Scholar]
  29. Talmazov, G. , Bencharit S., Waldrop T. C., and Ammoun R.. 2020. “Accuracy of Implant Placement Position Using Nondental Open‐Source Software: An in Vitro Study.” Journal of Prosthodontics 29, no. 7: 604–610. 10.1111/jopr.13208. [DOI] [PubMed] [Google Scholar]
  30. Terrabuio, B. R. , Carvalho C. G., Peralta‐Mamani M., Santos P. S. d. S., Rubira‐Bullen I. R. F., and Rubira C. M. F.. 2021. “Cone‐Beam Computed Tomography Artifacts in the Presence of Dental Implants and Associated Factors: An Integrative Review.” Imaging Science in Dentistry 51, no. 2: 93. 10.5624/isd.20200320. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Xu, Z. , Zhou L., Zheng M., et al. 2023. “The Effect of Implant Surgery Experience on the Learning Curve of a Dynamic Navigation System: An In Vitro Study.” BMC Oral Health 23, no. 1: 89. 10.1186/s12903-023-02792-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Yang, F. , Chen J., Cao R., et al. 2024. “Comparative Analysis of Dental Implant Placement Accuracy: Semi‐Active Robotic Versus Free‐Hand Techniques: A Randomized Controlled Clinical Trial.” Clinical Implant Dentistry and Related Research 26, no. 6: 1149–1161. 10.1111/cid.13375. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Yang, F. , Chen J., Wang L., and Ding Y.. 2024b. “Accuracy and Safety of Implant Placement With a Novel Semi‐Autonomous Robotic‐Assisted Surgical System: A Translational Research Study.” Journal of Prosthetic Dentistry, ahead of print, November 22. 10.1016/j.prosdent.2024.10.026. [DOI] [PubMed] [Google Scholar]
  34. Yu, M. , Luo Y., Li B., Xu L., Yang X., and Man Y.. 2024. “A Comparative Prospective Study on the Accuracy and Efficiency of Autonomous Robotic System Versus Dynamic Navigation System in Dental Implant Placement.” Journal of Clinical Periodontology 52, no. 2: 280–288. 10.1111/jcpe.14091. [DOI] [PubMed] [Google Scholar]
  35. Yu, X. , Tao B., Wang F., and Wu Y.. 2023. “Accuracy Assessment of Dynamic Navigation During Implant Placement: A Systematic Review and Meta‐Analysis of Clinical Studies in the Last 10 Years.” Journal of Dentistry 135: 104567. 10.1016/j.jdent.2023.104567. [DOI] [PubMed] [Google Scholar]
  36. Zhang, S. , Cai Q., Chen W., et al. 2023. “Accuracy of Implant Placement via Dynamic Navigation and Autonomous Robotic Computer‐Assisted Implant Surgery Methods: A Retrospective Study.” Clinical Oral Implants Research 35, no. 2: 220–229. 10.1111/clr.14216. [DOI] [PubMed] [Google Scholar]
  37. Zhao, X. Z. , Xu W. H., Tang Z. H., Wu M. J., Zhu J., and Chen S.. 2014. “Accuracy of Computer‐Guided Implant Surgery by a CAD_CAM and Laser Scanning Technique.” Chinese Journal of Dental Research 17, no. 1: 31–36. [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Figure S1. Workflow of patients’ enrolment.

CLR-36-725-s001.docx (97.4KB, docx)

Data Availability Statement

The datasets used and/or analyzed in the current study are available from the corresponding author upon reasonable request.

Articles from Clinical Oral Implants Research are provided here courtesy of Wiley

RESOURCES