VR-based annotation assistance system for volumetric medical image segmentation

Mika Anttonen, Dongwann Kang

Research output: Contribution to journalArticlepeer-review

Abstract

The application of Virtual Reality (VR) in the healthcare sector has been increasingly recognized, particularly in the realms of medical imaging and clinical skill training. Despite the variety of software and tools available for medical imaging on desktop platforms, their VR counterparts often lack advanced features, notably in the area of annotation tools. Existing VR annotation software, while capable of accelerating the annotation process, is hampered by framerate issues. In our work, we introduce two novel methods for annotating medical images in VR, designed to expedite the annotation process while maintaining a satisfactory framerate. Leveraging VR's capability to display medical images in three dimensions, our approach accelerates the area selection process and sustains a higher framerate by reducing the number of objects generated during annotation visualization. Our experiments revealed that our proposed method not only matches the speed of existing VR annotation software but also significantly reduces or even eliminates framerate problems, varying with the visualization technique implemented.

Original languageEnglish
Pages (from-to)1209-1222
Number of pages14
JournalKSII Transactions on Internet and Information Systems
Volume19
Issue number4
DOIs
StatePublished - Apr 2025

Keywords

  • Hounsfield Unit-Based ROI Selection
  • Medical Image Annotation
  • Medical Image Segmentation
  • Virtual Reality
  • Volumetric Model Visualization

Fingerprint

Dive into the research topics of 'VR-based annotation assistance system for volumetric medical image segmentation'. Together they form a unique fingerprint.

Cite this