Clinical validation of an automated zygomatic implant planning system: An international multicenter study

Abstract

Objective: To validate the performance of a novel automated zygomatic implant (ZI) planning tool (Zygoplanner) in an international multicenter cohort. Methods: Preoperative computed tomography (CT) scans from 69 patients (276 ZIs) who underwent quad zygomatic implant surgery were retrospectively analyzed using Zygoplanner. Automated ZI plans generated by Zygoplanner were compared with the corresponding clinically placed implants. The comparison metrics included three-dimensional bone-to-implant contact (3D-BIC), Zygoma Anatomy-Guided Approach (ZAGA) classification, safety distances to the orbital rim and the inferior border of the zygoma, inter-implant spacing, surgeon subjective assessments, and a blinded Turing test. Results: Compared with clinically-proved manual plans, Zygoplanner-based automated ZI plans achieved significantly higher 3D-BIC values (p < 0.001). In the internal dataset, automated plans showed slightly shorter orbital rim distances (4.51 vs 5.71 mm, p = 0.022) but significantly greater distances to the inferior zygomatic border (8.21 vs 7.20 mm, p = 0.015). In the external dataset, Zygoplanner demonstrated significantly larger inter-implant spacing (3.01 vs 1.18 mm, p < 0.001). Despite these differences, ZAGA classifications remained consistent between both groups. Subjective evaluations by experienced surgeons yielded high scores for stability (4.63/5), feasibility (4.51/5), and safety (4.25/5). In the blinded Turing test, Zygoplanner-generated plans were identified as human expert in 65.2 % of cases, significantly exceeds the 50 % random guessing threshold (p = 0.0075). Conclusion: Zygoplanner automatically generates clinically acceptable quad ZI plans with improved 3D-BIC, which effectively reduces the planning complexity and facilitates better preoperative ZI planning. Clinical Significance: Zygoplanner shows strong potential to automate and standardize quad zygomatic implant planning, thereby improving surgical predictability and safety while simplifying the planning workflow. Its automated workflow supports more consistent outcomes and may broaden access to advanced zygomatic implant rehabilitation for patients with edentulous maxillae.