J. C. Rajapakse, F. Kruggel, J. M. Maisog, and D. Von Cramon, "Modeling hemodynamic response for analysis of functional MRI time-series, " Hum. Brain Mapp., vol. 6, pp. 283--300, 1998.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....can be reliably estimated prior to analysis. Typically, the time course of stimulus presentation or task variation is convolved with a gamma, Poisson or Gaussian response kernel, or else a combination of Fourier series are used to generate one or more expected hemodynamic response (HR) functions [11 13]. The actual time courses of every voxel or smoothed voxel region are then compared to the selected template(s) and statistical models are used to identify regions whose time courses are significantly correlated to the models and to determine the magnitudes of their model related activation ....

Rajapakse, J. C., Kruggel, F., Maisog, J. M., von Cramon D. Y. 1998: Modeling hemodynamic response for analysis of functional MRI time-series. Hum. Brain Mapping 6: 283-300.

....is to model the temporal response as a linear, stationary system (Lange and Zeger, 1997; Department of Theoretical Statistics, Department of Mathematical Sciences, University of Aarhus, Ny Munkegade, DK 8000 Aarhus C, Denmark. e mail: vaever imf.au. dk III.1 Friston et al. 1994; Cohen, 1997; Rajapakse et al. 1998). Assuming that neural activation follows the stimulation function, the convolution of the latter with the impulse response function (or haemodynamic response function) then completely describes the temporal response. This seems to be reasonable in many experiments (Cohen, 1997; Dale and Buckner, ....

Rajapakse, J.C. et al. (1998) Modeling hemodynamic response for analysis of functional MRI time-series. Human Brain Mapping, 6, 283-300.

.... magnetic resonance imaging (fMRI) and positron emission tomography (PET) are popular modalities to image the working human brain [4] 5] Functional brain studies acquire a series of head scans while the subject is alternatively performing a major sensory or cognitive task and a baseline task [6], where the input stimulus to the brain takes the form of an ON OFF box car pattern. The first step in the analysis of functional brain images is to detect brain regions that are activated by input stimuli [5] This may be seen as a classification or segmentation of brain voxels into activated ....

....intensity variation and subject s head motion. Derivation of SPMs is usually the first step in the statistical analysis of functional images, in order to detect regions of significant activation. In what follows the analysis of SPMs is presented in the framework of the general linear model (GLM) [6], 19] We let denote the set of brain voxels. A. General Linear Model (GLM) Consider an fMRI experiment involving multiple input stimuli, and let and denote the values of the fMRI time series response and the input stimulus at time , respectively. Let and the design matrix of the experiment be ....

[Article contains additional citation context not shown here]

J. C. Rajapakse, F. Kruggel, J. M. Maisog, and D. Y. von Cramon, "Modeling hemodynamic responses for analysis of functional MRI time-series, " Human Brain Map., vol. 6, no. 4, pp. 283--300, 1998.

....slices for visualization. 5.1. Visual Experiment An 8KHz alternating checker board pattern with a central fixation point was projected on an LCD system, and subjects were asked to fixate on the point of stimulation. FLASH images at three axial levels of the brain at the visual cortex were taken [17]. The detected activation using the SPM, SICA and ICA R are shown in the Fig. 2, and Fig. 3 shows the time series corresponds to the activation pattern detected from the first axial slice using ICA R technique. 5.2. Memory Retrieval Task The subjects performing the experiment learned three ....

J. C. Rajapakse, F. Kruggel, J. M. Maisog, and D. Y. von Cramon, "Modeling hemodynamic response for analysis of functional MRI time-series," Human Brain Mapping, vol. 6, pp. 283--300, 1998.

....one may only want to reliably obtain a particular desired component or a set of desired sources, and automatically discard the rest of uninteresting signals or noises. At some instances, a trace of the desired signals is available, for example, the On Off stimulation scheme of an fMRI experiments [10]. In this section, a variation to the classical ICA is proposed, in which only a set of desired independent components (ICs) is extracted incorporating the prior information as reference signals; we refer this technique as ICA with reference. These reference signals carry some information of the ....

....desired signal , respectively, e) and (f) the reference and the output for extracting the desired signal (g) and (h) the outputs extracted by using the references of , respectively. 4.2. Synthetic fMRI Time Series Data Multiple input stimuli have often been used in fMRI experiments [10]. FMRI time responses from activated brain voxels are always confounded by physiological signals such as cardiac, respiratory, and blood flow, and the electronic noise of the scanners [10] To simulate this situation, two fMRI time responses from activated voxels # ) ....

[Article contains additional citation context not shown here]

J. C. Rajapakse, F. Kruggel, J. M. Maisog, and D. Y. von Cramon, "Modeling hemodynamic response for analysis of functional MRI time-series," Human Brain Mapping, vol. 6, pp. 283--300, 1998.

....representing the stimuli with a model function for the HR. A number of functions have been proposed for this purpose: Friston et al. 5] used a global Poisson function whose parameter was estimated from the data. Lange and Zeger [6] instead proposed to use the Gamma function and Rajapakse et al. [7] the Gaussian function. In both of the latter cases the model function parameters were estimated separately for each pixel averaged over data from many trials in the Fourier domain. Friston et al. 8] proposed modeling the HR using a linear combination of fixed global basis functions chosen to ....

....function, relative to some baseline level. The function can be fitted to a set of data by minimizing, e.g. the least squares fit with respect to the parameters, using standard methods for nonlinear optimization (see, e.g. 12] Note that, in comparison to the model proposed by Rajapakse et al. [7], this model does not involve a convolution with a function representing the stimulus. This is equivalent to representing the stimulus as an impulse at the beginning of each trial, which will be a reasonable approximation as long as the stimulus is short relative to the temporal resolution of the ....

J. C. Rajapakse, F. Kruggel, J. M. Maisog, and D. Y. von Cramon, "Modeling hemodynamic response for analysis of functional MRI time-series, " Human Brain Mapping, vol. 6, pp. 283--300, 1998.

No context found.

Rajapakse, J.C., Kruggel, F., Moisag, J., and von Cramon, D.Y. (1998). Modeling hemodynamic response for analysis of functional MRI time-series, Human Brain Mapping, 6, 283-300.

No context found.

J. C. Rajapakse, F. Kruggel, J. M. Maisog, and D. Von Cramon, "Modeling hemodynamic response for analysis of functional MRI time-series, " Hum. Brain Mapp., vol. 6, pp. 283--300, 1998.

No context found.

Jagath C. Rajapakse, Frithjof Kruggel, Jose M. Maisog, and D. Yves von Cramon. Modeling hemodynamic response for analysis of functional MRI time-series. Human Brain Mapping, 6:283-300, 1998.

No context found.

J. Rajapakse, F. Kruggel, J. Maisog, and v. Y. D. Cramon, "Modeling hemodynamic response for analysis of functional MRI time-series," Human Brain Mapping 6, pp. 283--300, 1998.

No context found.

Rajapakse, J.C., Kruggel, F., Maisog, J.M., and von Cramon, D.Y. (1998). Modeling hemodynamic response for analysis of functional MRI time-series. Human Brain Mapping, 6:283-300.

No context found.

Rajapakse, J. C., F. Kruggel, J. M. Maisog, and D. Y. von Cramon (1998). Modeling hemodynamic response for analysis of functional MRI time-series. Human Brain Mapping 6 (4), 283300.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC