Canonical text (the exact bytes fingerprinted)

TRUVACE RECORD VERSION
record: TRV-2026-2323
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-10-31T18:34:34.416031Z
status: published
lens: g_space
sector: health
headline: AI-assisted versus point-based registration for digital planning in CBCT images with metallic artifacts
dek: Accurate registration between cone-beam computed tomography (CBCT) images and intraoral scans is fundamental to digital implant planning because registration errors may propagate throughout the digital workflow and affect the accuracy of computer-guided implant surgery. Conventional point-based registration relies on manually selected anatomical landmarks and may be compromised by metallic artifacts that obscure dental structures. Artificial intelligence (AI)-assisted registration has recently emerged as a promi…
gain_title: In 45 patients with metal-distorted dental scans, AI-assisted registration aligned images with a median discrepancy of 0.190 mm versus 0.268 mm for manual point-based registration, with no added time.
problem_title: (none)
trace_subject: (none)
gain_reading: In 45 patients with metal-distorted dental scans, AI-assisted registration aligned images with a median discrepancy of 0.190 mm versus 0.268 mm for manual point-based registration, with no added time.
gain_evidence: AI-assisted registration demonstrated significantly lower registration discrepancies than conventional point-based registration (median 0.190 mm [IQR 0.121-0.275] vs. 0.268 mm [IQR 0.200-0.418]; p < 0.001)
problem_reading: (none)
problem_evidence: (none)
quick_read: Researchers compared two ways of aligning cone-beam CT images with intraoral scans for digital implant planning in patients whose CT images contained metallic artifacts. One method used AI assistance and the other relied on manually chosen anatomical landmarks.

The AI-assisted approach produced smaller alignment errors across all artifact severity levels while taking about as long. This matters because alignment errors can carry through the planning workflow into guided implant surgery, but the authors say multicenter, multi-software, prospective confirmation is still needed.
limitation: The authors state that further prospective multicenter studies using multiple implant-planning software platforms and broader clinical scenarios are needed to confirm the findings.
tag: Evidence-backed gain
key_points: The study included 45 patients and 78 paired arch observations analysed for accuracy. | Median discrepancy was 0.190 mm with AI assistance versus 0.268 mm with point-based registration (p < 0.001). | Errors rose with artifact burden for both methods, but AI-assisted registration stayed lower in mild, moderate and severe groups. | Registration time was comparable, and intra-observer reliability was excellent (ICC 0.985).
rundown: Accuracy was measured at the arch level as the mean discrepancy at three predefined anatomical landmarks using Geomagic Control X software. Arches were grouped by artifact burden into mild, moderate and severe categories.

Because the data were not normally distributed, the team used paired Wilcoxon signed-rank tests. A generalized estimating equations sensitivity analysis accounted for upper and lower arches within the same patient and confirmed the main result.

The authors conclude that AI-assisted registration may improve the accuracy and consistency of multimodal image registration in implant planning, particularly in difficult cases.
sources:
- peer_reviewed | BMC Oral Health | https://doi.org/10.1186/s12903-026-09990-0 | 2026-10-31
prev: 0000000000000000000000000000000000000000000000000000000000000000

sha256

28de6c62bd17eadc4dad7516352032cb34fcb46ac4b5b6bc7948e13a1c7cbbc0

previous

0000000000000000000000000000000000000000000000000000000000000000