Medical Image Analysis Lab

Photo Gallery

Figure 3.2: Model for applying ICA to fMRI data

Figure 3.1: Comparison of GLM and ICA (left) and ICA illustration (right). The GLM (top left) is by far the most common approach to analyzing fMRI data, and to use this approach, one needs a model for the fMRI time course whereas in spatial ICA (bottom left), there is no explicit temporal model for the fMRI time course, this is estimated along with the hemodynamic source locations. (right) The ICA model assumes the fMRI data, x, is a linear mixture of statistically independent sources, s and the goal of ICA is to separate the sources given the mixed data and thus determine the s and A matrices

Figure 3.7: (left) Mean activation maps from patients with schizophrenia and healthy controls. Right auditory cortex demonstrated the greatest difference (white box); (middle) Right auditory cortex difference maps with optimized boundaries; (right) Individual classification results for cohort 1, and replication in cohort 2. Schizophrenia classification is indicated with the color red.