FMRI, or Functional Magnetic Resonance Imaging, is a technique for determining which parts of the brain are activated by different types of physical sensation or activity.
Abstract
FMRI, or Functional Magnetic Resonance Imaging, is a technique for determining which parts of the brain are activated by different types of physical sensation or activity, such as sight, sound or the movement of a subject’s fingers. This “brain mapping” is achieved by setting up an advanced MRI scanner in a special way so that the increased blood flow to the activated areas of the brain shows up on Functional MRI scans. fMRI is becoming the diagnostic method of choice for learning how a normal, diseased, or injured brain is working, as well as for assessing the potential risks of surgery or other invasive treatment of the brain. In order to extract the sources of the brain activity from the experiment results, BSS algorithm can be used. The blind source separation problem is to extract the underlying source signals from a set of linear mixtures, where the mixing matrix is unknown.
The problem (or the background)
After fmri experiment we get some series of the frames (images), which represent a brain activity during the experiment. The problem is, that the sources are dependent, and therefore it’s hard to locate exactly the activated regions in the image. BSS approach could be used in this case but the existing BSS algorithms are designed to deal mainly with time series.
The solution (or the basic approach)
Lately in the research of Dr. Tzibulevski was found that existent BSS algorithms can be used if sparse transformation is being applied to the mixed pictures previously. Therefore, the solution consists of two stages:
1) applying sparse transformation to the mixed pictures
2) transforming the pictures to vectors and using the existing BSS approaches (Infomax or JADE)
Tools
The Project was programmed in Matlab 6.5, on a PC platform. Matlab tools used are the Wavelet toolbox, image processing, PCA tools. The ICA toolbox, developed by Scott Makeig of the Salk Institute, was used to implement Infomax separation.
Conclusions
– Independent sources of the brain activity were successfully extracted from the series of the fMRI images.
– The only way to estimate the quality of the obtained results is by using some fMRI analysis tool – some medical background is needed. Maybe in future project will be proposed to compare the results of blind source separation obtained in the project to the existing tools results.
Results
Acknowledgment
We are grateful to our project supervisor Michael Bronstein for his guidance. We would also like to thank Johanan Erez, Dr.Michael Tzibulevski and the rest of the Image Science lab staff, for their support throughout this work.
We are also grateful to the Ollendorf Minerva Center Fund for supporting this project.