COMPRESSIVE BRAIN NEURAL ACTIVITY DETECTION USING FUNCTIONAL MAGNETIC RESONANCE IMAGES by

Functional Magnetic Resonance Imaging (fMRI) measures the hemodynamic re-sponse related to neural activity in the brain or spinal cord of human subjects. The goal of our research is to develop a framework of data processing and logic inference which can identify the neural activity patterns using fMRI measurements with high accuracy at low computational cost. Throughout the development of our fMRI brain neural activity detection/identification framework, the following issues have been focused and investigated: 1) Selection and modeling of a priori. The prior knowledge in the context of fMRI data analysis includes the sparsity of neural activity, the spatial, tem-poral, inter-subject correlations and structural invariance of fMRI measure-ments. Proper mathematical models of prior knowledge should be developed and utilized to reduce the data volume, restore image distortions, and remove false alarms, which will lead to higher detection accuracy and lower process-ing cost. We have explored and developed a series of a priori models rangin