Virtual Reality in Clinical Teaching and Diagnostics for Liver Surgery: Prospective Cohort Study

Learning and applying anatomy are essential but are studied and done through 2D tools and imaging techniques. This study aims to verify the usefulness of an additional 3D technique and ensure an improvement in the visualization of anatomical structures and pathological findings.

The study aims to examine the usefulness of virtual reality (VR) technology as an additional tool in medical diagnostics. Groups of students, residents, and specialists in surgery, radiology, and internal medicine evaluated magnetic resonance imaging (MRI) by answering a multiple-choice questionnaire. Subsequently, a virtual 3D display was used for processing. The questionnaire focused on topographical conditions and the transfer of academic knowledge into clinical application. The main objective was to determine anatomical understanding in a comparison between sectional image (2D) presentation and additional VR (3D) presentation, measured through correctly answered questions and processing time. The system usability scale (SUS) was integrated as another criterion for VR usability.

The cross-over study assessed 63 participants regarding their knowledge of liver anatomy and pathologies based on an interindividual comparison. Group formation according to the respective level of medical training was as follows: students (n=35), residents (n=15), and specialists (n=13). Participants answered 25 multiple-choice questions first using sectional imaging (MRI) in a 2D environment (computer screen) and afterward with the respective segmented 3D model visualized in a VR simulation. The main criteria for the analyses were the number of correctly answered questions and processing time. A customized SUS was used to analyze VR usability. Missing data analysis showed that there were no accounted missing data.

The rate of correct answers improved significantly with the additional use of VR (F _1,59=314.376; P<.001). Using MRI, a significant difference was observed between students and residents (P=.04) and between students and specialists (P<.001). In the VR condition, no significant differences between groups were found. In the MRI condition, significant differences in processing time were observed between students and specialists (P=.02) and between residents and specialists (P=.04). No differences existed between students and residents. With VR, processing time decreased significantly in all groups (F _1,59=280.700; P<.001). Significant differences between students and specialists (P=.02) and between students and residents (P=.004) remained. No notable differences between residents and specialists (P=.72) were found. The SUS showed a subjectively simplified answerability of the questions with additional use of VR. The usefulness and benefits for an additional use of VR were stated.

The additional use of VR suggests statistically significant improvements across all groups. VR seems to enable students and residents to participate in diagnostics and create treatment plans at an early stage. Transferred to clinical practice, this may lead to improvement in diagnostics and interventions. The lack of randomization and a potential learning effect are the main limitations to be addressed in future studies.