On the use of temporal correlation coefficients for magnetic resonance mapping of functional brain activation: Individualized thresholds and spatial response delineation
✍ Scribed by Andreas Kleinschmidt; Martin Requardt; Klaus-Dietmar Merboldt; Jens Frahm
- Publisher
- John Wiley and Sons
- Year
- 1995
- Tongue
- English
- Weight
- 796 KB
- Volume
- 6
- Category
- Article
- ISSN
- 0899-9457
No coin nor oath required. For personal study only.
✦ Synopsis
Abstract
Functional activation of the human visual and motor system was studied by magnetic resonance imaging (MRI) at 2.0 T using dynamic series of oxygenation‐sensitive gradient‐echo images at high spatial resolution. Activation maps were computed by correlating signal intensity time courses with a reference waveform on a pixel‐by‐pixel basis. Although this strategy readily demonstrates stimulus‐related functional cooperativity of activated regions in thresholded maps of correlation coefficients, intertrial variability in the underlying distributions of correlation coefficients precludes the use of correlation coefficients as direct thresholds for defining activation. Because stimulus‐related effects emerge as positive deviations from an otherwise symmetric distribution of correlation coefficients, invariance against intertrial differences as well as adequate visualization of activated areas may be achieved by the following procedure. First, a symmetrized noise distribution is reconstructed from the actual activation map that allows rescaling of correlation coefficients into percentile ranks with respect to the integral of the noise distribution. Second, a high percentile rank (or correspondingly low error probability) can be used as threshold to define primary sites of activation with high specificity. And third, the spatial extent of activation may be delineated by adding directly neighboring pixels with lower values provided their correlation coefficients are high enough to contribute to the positive deviation from the noise distribution. The outlined approach yields robust activation maps but still awaits a more thorough statistical treatment of activation in MRI correlational mapping.