Wavelet Transform Based Accurate Estimation of Hemodynamic Response Function in functional MR images

Based on principals of blood oxygenation level dependent (BOLD) effect, MR signals raised according to brain neural activity. Such temporal change of MR signal is called Hemodynamic response function (HRF). Because HRF varies among activation sites and among subjects, we have been considering that estimating HRF will be available for analyzing neurological condition of cerebrum and for investigating and diagnosing cerebral diseases. Many conventional methods for analyzing functional MR images have been proposed, however, there are few methods that can detect activation areas in the cerebral cortex and can estimate HRF. This paper proposes a method that detects activation areas and estimate HRF simultaneously. The performance of estimating HRF is evaluated by phantom study, and results are compared with statistical parametric mapping (SPM).