A real-time display methods for large-scale human body data

Byeong Seok Shin, Neil Y. Yen, Jong Hyuk Park, Young Sik Jeong

Research output: Contribution to journalArticlepeer-review

1 Scopus citations

Abstract

As the importance of big-data processing has increased, a considerable amount of research has been done in several areas. We focused on developing methods and applications of multimedia big-data manipulation in the medical field, particularly display methods for large-scale human body data. We proposed context-aware detail level selection and block-based volume rendering methods to generate high-quality images in real-time by handling the huge amount of data efficiently, even in stand-alone computers. We also proposed Gaussian distribution sampling and morphological erosion that produce good quality images even when we used roughly segmented volume data. An entire rendering system is implemented to make the best use of a graphics processing unit’s (GPU) capabilities for real-time processing. To verify the effectiveness of these algorithms, we developed two medical applications called virtual endoscopy and virtual dissection. The experimental results shows that our method efficiently provides high-quality medical visualization even when we use a large medical volume dataset.

Original languageEnglish
Pages (from-to)1-27
Number of pages27
JournalMultimedia Tools and Applications
DOIs
StateAccepted/In press - 9 Mar 2016

Keywords

  • Huge medical data visualization
  • Level-of-detail
  • Multi-volume rendering
  • Virtual dissection
  • Virtual endoscopy

Fingerprint

Dive into the research topics of 'A real-time display methods for large-scale human body data'. Together they form a unique fingerprint.

Cite this