Biomag96: Combining Modalities to Investigate Cortical Retinotopy

This poster was presented at the Tenth International Conference on Biomagnetism, held in Santa Fe, New Mexico, U.S.A., February 1996.

A short paper (ps.gz, 153K, 4 pages with 1 color picture) based on this poster was published in Biomag 96: Proceedings of the Tenth International Conference on Biomagnetism, Vol. II, Springer-Verlag, New York, pp. 1138--1141, 2000.



		COMBINING MODALITIES TO INVESTIGATE CORTICAL RETINOTOPY Monica K. Hurdal and D. L. S. McElwain Queensland University of Technology Department of Mathematics Brisbane, Queensland Australia



		*Introduction* The visual field has an ordered mapping onto the primary visual cortex (V1). More area of visual cortex represents central vision when compared to peripheral regions [1, 2]. Thus, the map between the visual cortex and the visual field changes as a function of retinal location. *Cortical Magnification* Cortical magnification (M) describes this retinotopic mapping and refers to an area of cortex that is devoted to representing a small area of visual field [3]. There is still much controversy regarding the precise details of cortical magnification and this retinotopic map. Thus, we want to investigate and estimate cortical magnification. A theoretical model is required to define the mapping. Use dipole source localization [4, 5] to investigate cortical magnification: model source of magnetic and / or electrical activity by a dipole given MEG or EEG data, find location of dipole which is best fit or best source that explains the observed data. Individual anatomy needs to be considered so a model of the visual cortex is also required. *The Theoretical Model* Represent cortex in Cartesian co-ords (x, y) and visual field in polar co-ords (r, ) Use retinal and cortical cell density and derivatives to define a novel mathematical formulation of the cortical mapping: small area in cortical plane = some small area in retina i.e. dx dy = M(r) r dr d and number of points in retina = number of points in cortex i.e. density in cortex * area = density in retina * area

The Cylindrical Model of the Primary Visual Cortex

Cortical features were obtained from T1 weighted MRI scans taken in a 1.5T field.

Scans were taken 1.5 mm apart with contiguous spacings of 1.5 mm.

Points located along the calcarine fissure were extracted from the MRI scans (Fig. 1).

A cubic spline curve was fitted to the data to model the calcarine fissure (Fig. 1).

The upper and lower visual cortex in one hemisphere of the brain are modeled as cylinders.

The cylinders lie beside each other and abut along the calcarine fissure.

Since the fissure is modeled by the spline curve, the cylinders also follow the shape of the spline curve.

The result is a 3D-cylindrical model of V1 which incorporates the realistic shape of the calcarine fissure.

Figures 2 and 3 show different orientations of this cylindrical model.

MRI slice
with calcarine fissure highlighted

Cubic spline curve fitted to
points representing the calcarine fissure

Figure 1: Points representing the calcarine fissure were extracted from MRI data. A cubic spline curve was then fitted to the data. The dimensions of the spline curve are with respect to a point in front of the head and are given in centimeters.


	Upper Visual Cortex

	Calcarine Fissure

	Lower Visual Cortex


	Upper Visual Cortex

	Calcarine Fissure

	Lower Visual Cortex

Figure 2: The primary visual cortex is modeled by two cylinders that follow the realistic shape of the calcarine fissure. The dimensions of the cylinders are with respect to a point in front of the head and are given in centimeters.

Figure 3: A different orientation of the cylindrical model.



		*The Next Step* A dipole source localization algorithm has already been coded and tested. In addition, the recording channel positions from MEG and EEG experiments can already be co-registered with the MRI data. Visual evoked potential (VEP) data has been collected from a checkerboard pattern reversal experiment and the neural sources estimated using the dipole source localization algorithm. The cylindrical model can now be used to relate the position of the estimated source to the position of the visual stimulus. In this manner, a truly human estimate for cortical magnification can be obtained. *Summary and Conclusions* A new and realistic model of the calcarine fissure and primary visual cortex has been developed. A formal and novel mathematical definition of cortical magnification has been obtained through new equations. Original method of combining modalities to investigate and estimate cortical retinotopy. New software and methodology with which to analyze VEP data. This methodology can be applied to other areas of brain research such as motor cortex and auditory cortex. *References* [1] Holmes, G. The organisation of the visual cortex in man, Proc. R. Soc. B., 1945, 132: 348-361. [2] Zeki, S. A Vision of the Brain, Oxford, Blackwell Scientific Publications, 1993. [3] Daniel, P.M., and Whitteridge, D. The representation of the visual field on the cerebral cortex in monkeys, J. Physiol., 1961, 159: 203-221. [4] Rush, S. and Driscoll, D. Current distribution in the brain from surface electrodes, Anesth. Analg., 1968, 47: 717-723. [5] Ary, J.P., Klein, S.A., and Fender, D.H. Location of sources of evoked scalp potentials: correction for skull and scalp thickness, IEEE Trans. Biomed. Eng., 28: 447-452. Copyright 1996, Monica K. Hurdal