Development Of A Mixed Reality–Based Organ Registration Assistance System For CT-Guided Puncture

Abstract

Extended Reality (XR) technologies have been increasingly applied in the medical field, particularly for supporting image-guided interventions. In CT-guided biopsy, accurate puncture requires repeated CT imaging to confirm the insertion path, which results in increased radiation exposure for both patients and operators. Mixed Reality (MR) has the potential to reduce this exposure by overlaying virtual anatomical information onto physical space; however, conventional MR-based alignment methods suffer from positional errors and limited operability when registering virtual objects to real organs. In this study, we propose an MR-based alignment method that reduces registration error by introducing two-dimensional positional markers overlaid on physical magnetic markers. The proposed approach aims to improve both alignment accuracy and usability when superimposing virtual objects onto real-world organ positions. An MR application was developed using Meta Quest 3 and Unity, enabling marker acquisition and virtual object placement through a combination of positional marker detection and controller-based interaction. To evaluate the effectiveness of the proposed method, we conducted a user study with 22 participants and compared two approaches: (1) visual manual placement of virtual markers and (2) automatic placement assisted by positional markers. Alignment accuracy, operation time, and subjective usability were assessed. Results showed that the proposed positional marker–assisted method significantly reduced positional error in the X and Y directions and shortened operation time compared to the visual placement method (p < 0.05). Subjective evaluations also indicated improved ease of operation and visual naturalness. These findings demonstrate that the proposed method effectively improves registration accuracy and usability in MR-based alignment for CT-guided biopsy support, suggesting its potential to contribute to safer and more efficient image-guided interventions.