ASA offers an extensive selection of neuroimaging methods for EEG and MEG analysis. Source modeling has advanced from multiple spatio-temporal dipole solutions to MUSIC, minimum norm estimate, LORETA, sLORETA, swLORETA and swLORETA in frequency domain - and ASA supports all of these. Moreover, realistic head modeling using boundary elements can be applied irrespective of what particular source modeling approach is used. MRI based individually shaped head models can be interactively generated and provide more accurate solutions.

ASA lets you apply source reconstruction methods within the regular workflow of your signal processing. All you need is good data!
First you load an individual head model or the standard model that is installed with ASA, electrode positions, and MRI, then you specify the time interval for analysis and before you know it you can run the source analysis that suits your ideas around this particular brain activation. Source results can be exported e.g. in Analyze format and compared with for instance fMRI data.

Individual MRI can be imported from DICOM, analyze (SPM) or Nifti files and is further processed for head and source modeling. The data are co-registered on the basis of anatomical landmarks. ASA provides interactive tools for the 3D image segmentation and head modeling, as well as a transformation from individual coordinates to standard Talairach coordinates.
more

ASA facilitates simultaneous fit of one or more current dipoles to the measured data in the least-squares sense. The Dipole Fit feature supports an automatic initial guess or an initial guess based upon another inverse solution (MUSIC). You are able to apply three different dipole models:

• The moving dipole model (all dipole parameters free over time),
  
• the rotating dipole (position fixed in time),
  
• and the fixed dipole (position and orientation stationary).
  

The multiple signal classification (MUSIC) is a probe dipole that is scanned through the source region. At each scan point its linear parameters are optimized. The field simulation from this locally optimal dipole is then projected onto the signal subspace of the data (based on singular value decomposition). The projection value, also referred to as the MUSIC metric, is displayed.

Brain activity, which is generated by distributed sources, can be reconstructed using the LORETA (Low Resolution Electromagnetic Tomography) algorithm. LORETA gives a solution that is spatially smooth. The time course of the activity of the distributed sources is overlaid as colored plots to 3D MRI images. LORETA can be applied to the analysis of evoked and continuous (raw) EEG and MEG data.

The sLORETA procedure allows you correct localization of human brain function for distributed sources. Compared to sLORETA the swLORETA feature exhibits a smaller localization error for deep sources and noisy signals.

The complexity of brain activity varties greatly, thus our mathematical modeling has to allow a sufficient degree of freedom to explain the measured data. ASA offers a representative set of methods for focal and distributed source modeling. Your results can be exported for further processing. The highly configurable 3D displays let you freely rotate and zoom into your results.