In vivo hemoglobin and water concentrations, oxygen saturation, and scattering estimates from near-infrared breast tomography using spectral reconstruction

Subhadra Srinivasan, Brian W. Pogue, Shudong Jiang, Hamid Dehghani, Christine Kogel, Sandra Soho, Jennifer J. Gibson, Tor D. Tosteson, Steven P. Poplack, Keith D. Paulsen

Research output: Contribution to journalArticle

60 Scopus citations

Abstract

Rationale and Objectives. Near-infrared (NIR) imaging has its niche in quantifying and characterizing functional changes in tissue relating to vascularity and metabolic status. Here, NIR tomography was applied to study mammographically normal breast tissue in vivo by evaluating relationships between functional parameters so obtained to clinical representers in an effort to understand factors influencing tissue compositional changes. Materials and Methods. A new spectral reconstruction method that is considered to provide the most accurate estimates of hemoglobin level, oxygen saturation, water fraction, scattering power, and amplitude was used to assess healthy breast tissue imaged in vivo by means of NIR tomography. The approach directly recovers functional parameters with inherent inclusion of spectral behavior enforced through the incorporation of a priori model assumptions. Sixty subjects were imaged by using a frequency-domain instrument followed by spectral image reconstruction and statistical analysis for significant correlations. Results. The new analysis shows statistically significant inverse correlations between body mass index and breast total hemoglobin and water fractions. Water fraction also correlated inversely with age and separated certain categories of breast density. Average scatter power was indicative of breast radiographic density composition, whereas scatter amplitude varied inversely with breast diameter. Total hemoglobin correlated with water fraction, whereas water correlated with scatter power. Conclusion. The changes observed here are attributable to volume fraction alterations and provide some of the most comprehensive data on breast composition variations with demographic factors.

Original languageEnglish
Pages (from-to)195-202
Number of pages8
JournalAcademic radiology
Volume13
Issue number2
DOIs
StatePublished - Feb 1 2006
Externally publishedYes

Keywords

  • Breast tissue characterization
  • Frequency-domain
  • Functional imaging
  • Near-infrared tomography

Fingerprint Dive into the research topics of 'In vivo hemoglobin and water concentrations, oxygen saturation, and scattering estimates from near-infrared breast tomography using spectral reconstruction'. Together they form a unique fingerprint.

  • Cite this