| |
Discussion and Future Work
183
Overall, a very fast computation was not the major goal because the user can do something
else while the program is fitting the atlases to the measured data. Useful functions and good
quality results, however, are essential to make sure that the tool has an advantage over the te-
dious and less systematic approach of manually marking the visual areas. Only then is it used
by many researchers. In this chapter, some recommendations are given on how the program
might be improved.
8.2 Atlases
·
If the subject does not fixate on the center of the stimulus, the atlases have to be ad-
justed. It would be helpful to have a function that creates atlases based on a given fixa-
tion point offset. Practically, this could be done by giving an additional parameter to
the function that creates the atlases. The fixation point offset could either manually be
entered or, even better, it could be tried to estimate it, e.g., based on the phase data. It
is not clear yet if this is possible in an unambiguous way.
·
If the user specifies more then 3 points, the additional points could be used not only
for the calculation of the energy, but also for the creation of the initial wedge-atlas.
Based on the additional points in various boundaries, the phase values along iso-
eccentricity lines could change with different slopes in different visual areas or even in
a non-linear way.
·
It would also be useful to be able to exclude regions in the atlases from the matching
procedure (by setting those regions in the atlases to NaN before the matching is done).
·
The ring atlas should be improved to take cortical magnification into account, i.e., the
phase should not change linearly over eccentricity.
·
It is already possible to create a wedge atlas that represents only one full hemifield.
This atlas can be used to identify only V1. However, it can also be used to identify any
other visual area that represents an entire hemifield, e.g., V3a. In addition, other at-
lases, for example for a half hemifield, could be created.
|  |
|
| |
|
|
