Unsupervised partial volume estimation in single-channel image data

Dzung L. Pham, Jerry L. Prince

Research output: Contribution to conferencePaper

Abstract

Partial volume effects are present in nearly all medical imaging data. These artifacts blur the boundaries between different regions, making accurate delineation of anatomical structures difficult. In this paper, we propose a method for unsupervised estimation of partial volume effects in single-channel image data. Based on a statistical image model, an algorithm is derived for estimating both partial volumes and the means of the different tissue classes in the image. To compensate for the ill-posed nature of the estimation problem, we employ a Bayesian approach that places a prior probability model on the parameters. We demonstrate on simulated and real images that the new algorithm is superior in several respects to the fuzzy and Gaussian clustering algorithms that have previously been used for modeling partial volume effects.

Original languageEnglish (US)
Pages170-177
Number of pages8
StatePublished - Jan 1 2000
EventMMBIA-2000: IEEE Workshop on Mathematical Methods in Biomedical Image Analysis - Hilton Head Island, SC, USA
Duration: Jun 11 2000Jun 12 2000

Other

OtherMMBIA-2000: IEEE Workshop on Mathematical Methods in Biomedical Image Analysis
CityHilton Head Island, SC, USA
Period6/11/006/12/00

    Fingerprint

ASJC Scopus subject areas

  • Analysis

Cite this

Pham, D. L., & Prince, J. L. (2000). Unsupervised partial volume estimation in single-channel image data. 170-177. Paper presented at MMBIA-2000: IEEE Workshop on Mathematical Methods in Biomedical Image Analysis, Hilton Head Island, SC, USA, .