Clinical Feasibility of <scp>AI</scp> ‐Driven Automated Virtual Dental Implant Placement: A Cross‐Sectional Comparative Study
Bibliographic record
Abstract
OBJECTIVES: To assess the clinical validity of artificial intelligence (AI)-driven virtual implant placement compared to human intelligence (HI)-based virtual and actual single implant placement in the posterior mandible. MATERIAL AND METHODS: Thirty-two patients for whom experts performed single implant placement in the posterior mandible were included, each with preoperative and postoperative cone-beam computed tomography (CBCT) scans. The preoperative scans were registered to the corresponding postoperative scans and used for both AI- and HI-driven virtual implant planning at the implant site. From each case, three implants' scenarios (i.e., HI-placed, AI-planned, and HI-planned) were exported and compared. The analysis focused on angular deviation and the spatial relationship of each implant to adjacent anatomical structures and the expert-designed prosthetic wax-up. In addition, pairwise comparisons were performed to quantify angular and linear deviations at both the coronal and apical levels. Implant length and diameter from planned versus placed implants were evaluated, and planning time and consistency were compared between AI- and HI-based approaches. RESULTS: AI-based planning showed no statistically significant differences compared to HI-based methods observed in angular deviation relative to adjacent tooth (HI-placed: 7.7° ± 5.6°, AI: 6° ± 4.7°, HI-planned: 5.2° ± 5.7°) and coronal deviation (AI vs. HI-placed: 0.9 ± 0.8 mm, AI vs. HI-planned: 0.8 ± 0.4 mm, HI-planned vs. HI-placed: 1.0 ± 1.1 mm), all with p > 0.05. Implant diameter and length were also consistent across the different approaches, with HI-placed (4.3 ± 0.3 mm; 9.7 ± 1.3 mm), AI (4.3 ± 0.4 mm; 9.9 ± 1.2 mm), and HI-planned (4.3 ± 0.4 mm; 9.8 ± 1.3 mm) showing no significant differences (p > 0.05). However, AI planning was significantly faster (36.3 ± 7.3 s vs. 373 ± 113 s) and more consistent, with a median surface deviation of 0 mm compared to 0.39 mm for HI (p < 0.05). CONCLUSION: The AI tool showed clinically valid implant selection, matched expert placement and planning in virtual implant positioning for missing mandibular premolars and molars while being highly consistent and 10 times faster compared to human expert planning.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.008 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".