Virtual-fit assessment of full-arch implant photogrammetry accuracy in a human cadaver model

To evaluate the accuracy of extraoral (EPG) and intraoral photogrammetry (IPG) based complete-arch implant workflows in the maxilla and mandible using a fresh human cadaver model, and to compare conventional best-fit deviation metrics with a clinically interpretable virtual-fit assessment.

An edentulous maxilla and mandible of a fresh human cadaver received six implants per arch, restored with multi-unit abutments; one low-torque implant was excluded. A reference dataset was obtained using an ATOS Q industrial scanner. Test scans were acquired with iCam, MicronMapper, and an Aoralscan Elite intraoral scanner using ioConnect scan bodies and/or Elite Photogrammetry markers. After conversion to multi-unit connection geometries, accuracy was evaluated using penetration-type best-fit alignment (mean absolute deviation, MAD) and a non-interpenetrating virtual-fit protocol, simulating prosthetic seating, and quantifying marginal gap (MG) and absolute marginal discrepancy (AMD).

In the mandible, iCam and MicronMapper demonstrated significantly better trueness (lower MG and AMD) than ioConnect, whereas in the maxilla, only iCam remained significantly superior. Precision was higher for iCam and MicronMapper than for ioConnect and IPG in the mandible. In the maxilla, iCam, MicronMapper, and ioConnect all showed significantly better precision than IPG. Median AMD ranged from 38 to 97 µm across systems, although individual IPG scans frequently exceeded 100 µm. MAD correlated strongly with MG and AMD (r = 0.89-0.91) but underestimated misfit, with discrepancies reaching up to 150 µm.

EPG demonstrated higher and more consistent accuracy than intraoral workflows. Virtual-fit assessment provided a clinically more meaningful evaluation than conventional best-fit metrics, which systematically underestimated prosthetic misfit because penetration-type ICP alignment distributes errors across the dataset, potentially masking localized misfit.

For complete-arch implant impressions, EPG currently provides the most reliable accuracy. Intraoral workflows, particularly IPG, may show greater variability and should be used with caution. Clinically interpretable evaluation methods are essential to determine the true clinical acceptability of digital implant workflows.