[ Main ] [ Outline ] [ Introduction ] [ MRI Basics ] [ Cross-Relaxometry ] [ Results ] [ References ] [ Appendix ]

Abstract

With this project, we intend to quantitatively characterize and localize the human primary visual cortex. The primary technique that was used in our analyses was Magnetic Resonance Imaging (MRI). We will first give a brief overview of why current MRI techniques are inadequate for quantitative parameterization. We will then go over some of the basics of MRI to set a platform for the actual method that we used to derive our results, Cross-Relaxometry. Finally, we will present how our results were able to give quantitative parameters that were characteristic of the human visual cortex.

Motivation

This project tries to put quantitative parameters with absolute units on the human visual cortex. Currently, this type of parameterization has been performed using the available Magnetic Resonance Imaging (MRI) techniques. However, the conventional algorithms employed until now have not been able to put absolute units on the parameters that are characteristic of the different parts of the brain. This project tries to experimentally demonstrate the results of a recently published paper [1] that highlights a sequence which provides quantitative parameters with absolute units. These parameters are distinct for the different parts of the brain and hence can be used to get an enhanced localization of the primary visual cortex.

Primary Visual Cortex

The primary visual cortex (also known as the Striate Cortex or V1) lies at the back of the brain and is responsible for processing visual stimuli. It does so when the neurons in the cortex fire action potentials (electric signals) as the stimuli appear within their receptive fields. The receptive fields represent a small region within the entire visual field of the individual. The photoreceptors in the retina pick up the photons that enter the eye, transform them into electric signals, and the optic nerve carries these signals through the lateral geniculate nucleus (LGN) to the visual cortex. The signals that come to the visual cortex actually end at the line of Gennari that borders the cortex and acts as the input to the cortex. From there, V1 then acts as the major distributor of all visual information that reaches the cortical areas [2].

Figure 1. Outline showing the pathway of electric signals on the optic nerve to the visual cortex.

Figure 2. Picture of the brain showing the primary visual cortex, the lateral geniculate nucleus (LGN), and the optic nerve.

Line of Gennari

The line of Gennari (Figure 3) is a band of highly myelinated fibers in the cortical layer that borders the visual cortex. It is also what gives the striate cortex its name. The line of Gennari fibers act as an input to the visual cortex from the LGN. The borders of the visual cortex can be identified by seeing the areas where the line of Gennari disappears [3].

Note: Myelin is a fatty substance that covers the neuron fibers, protecting the neurons and helping in the fast transmission of electric signals along them. It also constitutes the well-known White Matter in the brain.

Figure 3. Horizontal section of the brain showing the line of Gennari in the striate cortex (primary visual cortex). From Polyzak (1957).

Magnetic Resonance Imaging

The basis for our project’s quantitative analyses was formed by Magnetic Resonance Imaging (MRI). MRI is used mainly in clinical and scientific studies to obtain high quality images of the brain and other parts of the body.

There are two types of MRI scans that are mainly used for brain scans:

Structural MRI

Structural MRI deals with information about tissue contrast. For example, in the brain, it helps distinguish between gray matter and white matter. Even though the structural MRI scan gives quantitative values for the different intensities observed, they are too close to distinguish sometimes. Thus if we look at the scan below in Figure 4, we have no positive way to identify the visual cortex or the Line of Gennari in this image. We can make a good guesstimate but cannot quantify it with any values.

(a) (b)

Figure 4. (a) Horizontal structural MRI scan of the brain. (b) Horizontal anatomical image of the brain showing the line of Gennari. It can be seen that (a) does not provide enough information to locate the line of Gennari in the scan.

Functional MRI (fMRI)

The functional MRI deals with determining which parts of the brain are activated by some physical or sensory stimulus. The stimulus can be applied easily while the person is lying in the scanner.

The scans obtained using fMRI are more informative since they can give a nice visual of the part that has been activated using the stimulus. For example, in Figure 5 we can see the yellow and red dots show visual cortex activity. It can be seen that the fMRI scan helps us to better localize the part of the brain in question. However, it suffers from the same fault as the structural case. fMRI does not work with units and hence it is not easy to quantify the different parameters of the brain.

Figure 5. An fMRI scan showing regions of activation, including the primary visual cortex.

(a) (b)

Figure 6. (a) A sophisticated fMRI scan outlining the primary visual cortex (V1) and other cortical regions (V2, V3, V4). (b) Horizontal anatomical image of the brain showing the line of Gennari. An image as that produced in (a) only gives relative values and hence it is not possible to quantitatively characterize the location of the visual cortex.

Project Proposal

As was outlined above, the line of Gennari borders the visual cortex. The line of Gennari as mentioned before is a highly myelinated band of fibers and if we are able to identify quantifiable methods which would outline this highly myelinated area and put absolute units on the values, we would have a successful localization of the visual cortex.

Hence, the project proposes to find a quantitative method that would produce certain parameters (called k and f parameters) that would be characteristic of the human visual cortex. The advantage here is that these parameters have different values in absolute units for different parts of the brain. Using these parameters, we should be able convincingly determine the location of the visual cortex.

Possible Applications

The method would have wide clinical and scientific applicability. It can be used to do an enhanced diagnosis of brain and white matter related diseases such as multiple sclerosis. This method would be able to quantitatively identify the affected areas and allow doctors to recommend better treatments that would be extremely localized and hence more effective.

Another application of this would be the quantitative parameterization of the visual cortex in blind individuals. In these individuals, it can hard to get fMRI data as was shown before since it would hard to activate the visual cortex area by giving them any visual stimulus. Hence, localization using fMRI techniques would not be very useful in these individuals. However, with this method would help solve this problem.