Publications

Recent advancements in computer technology have resulted in the rise of X-reality (XR) systems. Some institutions have begun to use XR systems as an alternative to cadaveric specimens and physical models. This transition has been carried out despite a lack of evidence about the efficacy of XR. In our earliest study, we compared the physical model to 3D, interactive projections on a 2D screen and found participants that learned from the physical model performed significantly better in nominal measures. Following this study, we explored the efficacy of more refined XR systems and established that participants who learned from the physical model performed significantly better in both nominal and functional measures of anatomical knowledge compared to those who learned from the Microsoft HoloLens, which is a mixed-reality (MR) device. In our current study, we explored the efficacy of the HTC Vive, a pure virtual-reality (VR) device, in comparison to the physical model in anatomical education. We hypothesized that given the enthusiasm surrounding this XR system, the VR model should perform at least as well as the physical model. We first conducted a preliminary qualitative study with 20 participants to develop an optimal learning environment for the VR model. Following this preliminary analysis, we recruited 20 McMaster University students with no prior formal anatomy education. Participants were allotted 10 minutes to learn 20 anatomical structures from a pelvic model on the VR headset. Participants were then given a 25-question test on a female, cadaveric pelvis with no time limit. The test questions were either nominal or functional. The nominal questions involved identifying the structures labelled on the cadaver and the functional questions involved interpreting the function of a structure based on its location and form. We compared the results of the VR participants to the 20 participants who learned on the physical model and the 20 participants who learned on the MR model from our HoloLens study. Our analysis demonstrated that participants learning on the physical model performed significantly better than the VR model when comparing total testing scores (56.4% vs 45.0%, respectively; p = 0.034). Furthermore, the VR model participants performed better than the MR model participants in terms of total testing score, although these results were not statistically significant (45.0% vs 37.6%, respectively; p = 0.267). In conclusion, these findings provide further evidence to support the superiority of the physical model over XR systems. Our future directions involve testing with other, more complex anatomical structures and exploring the factors that contribute to the superiority of the physical model, such as the role of stereopsis.

Support or Funding Information

Self-funded

Recent technological advancements in X-Reality (XR) seem to create remarkably realistic models for anatomical education. Despite a lack of evidence regarding the efficacy of XR in this context, several institutions have adopted these technologies as primary educational tools in anatomy as an alternative to traditional cadaveric laboratories. In our earliest study, we evaluated a 3D, interactive projection on a 2D screen against a physical model of a female pelvis. This data demonstrated that those who learnt on the physical model performed significantly better during testing. Subsequently, we explored the efficacy of more intricate XR systems. Thus, we compared the efficacy of the Microsoft HoloLens, a mixed-reality (MR) device, to a physical model in anatomy education. We recruited 20 McMaster University students and ran a preliminary study to gather qualitative data regarding the optimal MR environment. We gathered participant preference based on their experience observing several virtual objects against different coloured backgrounds and various lighting combinations. We used this data to build the testing environment for the MR model, such as adding black curtains and floor tiles to the room, and using a single light over the projection. These conditions were also used for the physical model, thus placing it at a slight disadvantage. We then recruited 40 McMaster University students with no prior anatomical education, and randomized them into two groups: one learning on a physical model of a female pelvis and one learning on the MR model of a female pelvis. We measured two possible covariates, spatial and stereoscopic ability, through two pretest assessments: a Mental Rotations Test (MRT) and a Titmus Fly Test, respectively. Our participants were then given 10 minutes to learn 20 structures using their respective models, and were tested on a female cadaveric pelvis on the basis of a 25-question test with no time limit. This test included 15 nominal questions, which asked participants to name the indicated structure, and 10 functional questions, which asked participants to determine the function of a structure based on its location and form. We hypothesized that due to the realistic model that the MR system created, it should perform at least equivalent to the physical model in the context of anatomical education. Our assessments found that participants learning on the physical model performed significantly better in comparison to their MR counterparts on both nominal (65% vs 41%, respectively; p = 0.0051) and functional measures (42% vs. 31%, respectively; p = 0.0134). Additionally, when controlling for the aforementioned covariates, we found these results to remain consistent. Ultimately, our results indicate that the MR device is an inefficient tool for anatomical education when compared to traditional physical models. Our future directions involve exploring possible determinants influencing the superiority of the physical model, such as stereoscopic vision, as well as the assessment of other XR systems, such as virtual reality headsets, in the context of anatomy education.

Support or Funding Information

Self-funded.