Potential of augmented reality technology in maxillofacial surgery: a literature review and results of clinical application

Background. The main characteristics of augmented reality (AR) and virtual reality (VR) technologies are immersion, presence, and interaction, which are defined by the technology used and individual perception. AR and VR automate processes and assist in areas that require repetitive tasks, particularly in medical education and training, including surgery. Recently, VR and AR have significantly entered the fields of maxillofacial surgery and dentistry, allowing doctors to create 3D models and conduct virtual surgeries, as well as train specialists on virtual models. These new methods require assessment of their usability and accuracy.

Aims — to evaluate the accuracy of our developed augmented reality system for creating surgical access in radicular cysts of the jaws.

Methods. We conducted a clinical comparative study to assess the accuracy of our developed augmented reality system using HoloLens software and Medgital Vision Editor for creating surgical access in radicular cysts of the jaws. Forty patients were selected and divided into three groups: Group 1 (n = 10) consisted of patients operated on using a surgical template; Group 2 (n = 20) consisted of patients operated on using a virtual template and augmented reality (AR); Group 3 (n = 10) consisted of patients operated on using the freehand method. In all three groups, preoperative computer modeling for surgical access, with a diameter of 5 mm, corresponding to a bone trephine, was performed based on CT scans of the jaws and intraoral scanning. Subsequently, all patients underwent surgical access to the cyst using a surgical trephine, followed by cystectomy and closure of the surgical wound. After the surgical intervention, all patients underwent a follow-up CT scan. The obtained CT data (DICOM files) were uploaded into Exoplan 3.0 software. We assessed the differences between the planned preoperative computer modeling and the actual surgical access performed. Furthermore, we analyzed the deviation angles of the formed surgical access and the depth preparation deviations. A statistical analysis of the obtained data was conducted.

Results. The analysis of the angle of deviation of the formed surgical access was 2.82, 2.25, and 9.77 in Groups I, II, and III, respectively. Significant differences were established (p < 0.001; method used — Kruskal–Wallis test). The analysis of the depth preparation deviation showed results of 0.53, 0.73, and 2.38 in Groups I, II, and III, respectively (p < 0.001; method used — Welch’s F-test). The accuracy results of our augmented reality navigation system are comparable to the accuracy achieved with surgical templates and significantly surpass the results obtained with the freehand method.

Conclusion. The accuracy of AR navigation is sufficient for clinical use, but some improvements are necessary.