Innovative visualization and segmentation approaches for telemedicine

D. Nguyen, Hans Roehrig, Marisa H. Borders, Kimberly A. Fitzpatrick, Janet Roveda

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In health care applications, we obtain, manage, store and communicate using high quality, large volume of image data through integrated devices. In this paper we propose several promising methods that can assist physicians in image data process and communication. We design a new semi-automated segmentation approach for radiological images, such as CT and MRI to clearly identify the areas of interest. This approach combines the advantages from both the region-based method and boundary-based methods. It has three key steps compose: coarse segmentation by using fuzzy affinity and homogeneity operator, image division and reclassification using the Voronoi Diagram, and refining boundary lines using the level set model.

Original languageEnglish (US)
Title of host publicationMedical Applications of Radiation Detectors IV
EditorsHans N. Roehrig, Lars R. Furenlid, H. Bradford Barber
PublisherSPIE
ISBN (Electronic)9781628412413
DOIs
StatePublished - 2014
EventMedical Applications of Radiation Detectors IV - San Diego, United States
Duration: Aug 18 2014Aug 21 2014

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume9214
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X

Other

OtherMedical Applications of Radiation Detectors IV
Country/TerritoryUnited States
CitySan Diego
Period8/18/148/21/14

Keywords

  • Fuzzy affinity
  • Large volume of image data
  • Segmentation
  • The Level Set model
  • The Voronoi Diagram

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Computer Science Applications
  • Applied Mathematics
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Innovative visualization and segmentation approaches for telemedicine'. Together they form a unique fingerprint.

Cite this