Biexponential diffusion decay in formalin-fixed prostate tissue : preliminary findings

Roger M. Bourne, Nyoman Kurniawan, Gary Cowin, Timothy Stait-Gardner, Paul Sved, Geoffrey Watson, Supriya Chowdhury, William S. Price

    Research output: Contribution to journalArticlepeer-review

    26 Citations (Scopus)

    Abstract

    Magnetic resonance microimaging was used to measure diffusion decay over an extended b-factor range in a formalin-fixed normal prostate sample and a Gleason pattern 3+4 cancer tissue sample. The coefficients of biexponential fits to diffusion decay data from 1600 voxels of dimension 160 x 160 x 160 µm3 in each sample were correlated with underlying epithelial and stromal compartment partial volumes estimated from high-resolution apparent diffusion coefficient (ADC) data (40 x 40 x 40 µm3 voxels) from the same tissue. In the normal tissue sample, the signal fractions of the low and high ADC components of the biexponential fits correlated linearly with partial volumes of epithelial tissue (R2 = 0.6) and stromal tissue (R2 = 0.5), respectively. Similar but weaker correlations were observed in the cancer sample. Epithelium-containing high spatial resolution voxels appeared to be composed of ~60% low ADC and ~40% high ADC component. Stromal voxels appeared to be composed of ~20% low ADC and ~80% high ADC component. This preliminary report suggests that distinctly different diffusion properties in microscopically adjacent cell types contribute to the multiexponential diffusion decay phenomenon in prostate tissue.
    Original languageEnglish
    Pages (from-to)954-959
    Number of pages6
    JournalMagnetic Resonance in Medicine
    Volume68
    Issue number3
    DOIs
    Publication statusPublished - 2012

    Keywords

    • apparent diffusion coefficient
    • cancer
    • diffusion decay
    • magnetic resonance imaging
    • prostate

    Fingerprint

    Dive into the research topics of 'Biexponential diffusion decay in formalin-fixed prostate tissue : preliminary findings'. Together they form a unique fingerprint.

    Cite this