Augmented reality for arthroplasty training - a randomised controlled trial of simulator-based training for acetabular cup orientation skills

Abstract

Background: Accurate implant orientation reduces wear and increases stability in arthroplast y. This is a technically demanding skill. Augmented Reality headsets overlay digital information over the real world. We have developed an enhanced AR headset capable of tracking bony anatomy in relation to an implant. Questions/Purposes: The purpose of t his study was to validate this enhanced AR headset in the training simulator setting as determined by (1) quantitative assessments accurately orientating an acetabular component, and (2) a questionnaire regarding the realism of the task, acceptability of the technology, and its potential role for surgical training. Methods: Twenty - four novice surgeons voluntarily participated in a simulation trial. Demographics and baseline performance in orientating an acetabular implant to six patient - specific values o n the phantom pelvis, were collected prior to training. All participants received standardised educational lectures based around four clinical vignettes on hip surgery to normalise their knowledge. Participants were randomised to two groups, completing fo ur, once - weekly sessions of baseline assessment, training, and re - assessment. One group trained using AR (with live holographic feedback of orientation), and the other received one - on - one tuition from an expert surgeon (ES). The solid - angle error in degree s between the planned and achieved orientations was measured. Participants completed baseline assessments, then training, and then assessment in this fashion after each weekly session, and a validated post - training questionnaire evaluating their experience s. Results: Novice surgeons of similar experience in both groups performed with a similar error prior to training ( Baseline ES mean error 16° ± SD 7° compared to AR 14±7°, p>0.05 ). During training, participants achieved average errors in orientations of 1±1° using AR, and 6±4° when guided by the ES (p<0.001). At final baseline assessment, average error was AR: 9±6° and ES: 8±5° (p>0.05). In post - training evaluation, 92% of participants would use the AR platform as a training tool for developing visuospati al skill, and 86% for procedure - specific rehearsal. 82% preferred ES for learning about orientation, but 58% preferred AR for practicing it. Conclusions: A novel head - mounted AR platform tracked an implant in relation to bony anatomy to a clinically - relev ant level of accuracy. It delivered training to novice surgeons for acquiring and retaining skills to orientate it, as effectively as an expert surgeon. In both groups, accuracy remained below ‘expert’ level proficiency (<5° error). Participants considered the AR platform feasible and useful as a training tool, with potential as an adjunct to learning in the operating theatre.