CAN MIXED REALITY BE AS ACCURATE AS PACS MEASUREMENTS IN ASSESSMENT OF GLENOID BONE LOSS?

Arthroscopic anatomic glenoid reconstruction (AAGR) for shoulder instability is of great interest to many surgeons. However, anatomical variations in acromion and coracoid can pose challenges for the procedure. Surgeons have traditionally used 2D-CT scans with 3D-reconstructions to better understand scapula anatomy. Even so, measuring on PACS software in 2D has limitations and inaccuracies, especially in determining where the coracoid and acromion are in 3D space. The advancements in 3D models and augmented reality can be utilized to better understand and quantify the anatomic variations in the scapula between patients. This study aimed to determine accuracy of measurements using a medical mixed reality (MR) system (RSQ-HOLO, RSQ Technologies, Poznan, Poland) compared to a commonly available PACS (IMPAX 6, Agfa, Mortsel, Belgium) in quantifying anatomic variations in the scapula. Glenoid height and width are simple and reproducible on PACS software and were therefore selected to compare against the MR system to determine the accuracy of measurements in 3D space. Based on sample size calculations, the CT scans of 80 patients requiring arthroscopic glenoid reconstruction were analyzed for glenoid size using the MR system. The height was measured as the maximum distance from the upper pole to the lower pole of the glenoid. The width was defined as the distance between the anterior and posterior glenoid rims perpendicular to and at 40% of the height. The same measurements were performed on the PACS system using the 3D reconstructed images. A second rater performed the same measurements on 40 patients, to determine the reliability of the measurements. The analysis had a cohort of 80 patients, with 45% being the right shoulder. Average age at time of CT scan was 39.7 +/- 18.3 years. The measurements conducted by both raters indicated mean differences of less than one millimeter in glenoid width and height when using the two different software applications. Specifically, in the measurements conducted by rater 1, the average height and width of the glenoid were 3.9±0.4cm and 2.8±0.5cm with the MR system, and 3.9±0.4cm and 2.8±0.4cm with the PACS system. Similarly, the measurements taken by the second rater were 3.9±0.3cm for glenoid height and 2.9±0.4cm for glenoid width with the MR system, and 3.9±0.4cm for glenoid height and 2.8±0.3cm for glenoid width with the PACS system (p>0.05). There was a strong level of agreement between the two raters, as evidenced by the Intraclass Correlation Coefficients (ICCs) exceeding 0.8. The MR measurements were consistent with traditional PACS measurements and the mean differences in measurements using two software systems are not clinically significant. This suggests that the MR system can serve as a viable alternative to the traditional PACS system for measuring CT scans. The MR system can be used to accurately measure anatomical differences in the scapula, where typical 3D reconstructed CT images fall short. This finding may lead to further advancements in personalized surgery utilizing 3D-modeling and mixed reality.