Comparison of different CT energy levels on liver fat prediction

Min Zhang, Suseon Yang, Tao Yang, Teresa Wu, Di Pan

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

Abstract

Computed Tomography (CT) is very popular for fatty liver diagnosis. Some studies demonstrate that the Hepatic Fat Content (HFC) is linearly correlated with Liver to Spleen ratio (L/S) of CT attenuation. However, L/S varies at different energy levels even for the same Region of Interest (ROI). We developed two Particle Swarm Optimization (PSO) Least Median Square (LMS) regression models in R (www.r-project.org) for single energy levels and multiple energy level combination respectively to evaluate their performance for determining and predicting the HFC levels. Monochromatic energy CT from 40keV to 140keV (11 levels) were performed on 3 patients with fatty livers and 4 ROIs were examined in this study. The results show that 80keV is the best for HFC prediction (Min Median Error Square, OF=7.46e-07) in single energy level model; and the combination of 60keV and 140keV performs the best (OF=7.198e-08) in multiple energy levels combination model.

Original languageEnglish (US)
Title of host publication61st Annual IIE Conference and Expo Proceedings
PublisherInstitute of Industrial Engineers
StatePublished - 2011
Event61st Annual Conference and Expo of the Institute of Industrial Engineers - Reno, NV, United States
Duration: May 21 2011May 25 2011

Other

Other61st Annual Conference and Expo of the Institute of Industrial Engineers
CountryUnited States
CityReno, NV
Period5/21/115/25/11

Keywords

  • CT Energy Level
  • Fatty Liver
  • Least Median Square Regression
  • Particle Swarm Optimization

ASJC Scopus subject areas

  • Industrial and Manufacturing Engineering

Fingerprint Dive into the research topics of 'Comparison of different CT energy levels on liver fat prediction'. Together they form a unique fingerprint.

  • Cite this

    Zhang, M., Yang, S., Yang, T., Wu, T., & Pan, D. (2011). Comparison of different CT energy levels on liver fat prediction. In 61st Annual IIE Conference and Expo Proceedings Institute of Industrial Engineers.