Using magnetic resonance imaging to assess visual deficits: a review

Abstract

Purpose: Over the last two decades, magnetic resonance imaging (MRI) has been widely used in neuroscience research to assess both structure and function in the brain in health and disease. With regard to vision research, prior to the advent of MRI, researchers relied on animal physiology and human post-mortem work to assess the impact of eye disease on visual cortex and connecting structures. Using MRI, researchers can non-invasively examine the effects of eye disease on the whole visual pathway, including the lateral geniculate nucleus, striate and extrastriate cortex. This review aims to summarise research using MRI to investigate structural, chemical and functional effects of eye diseases, including: macular degeneration, retinitis pigmentosa, glaucoma, albinism, and amblyopia.

Recent findings: Structural MRI has demonstrated significant abnormalities within both grey and white matter densities across both visual and non-visual areas. Functional MRI studies have also provided extensive evidence of functional changes throughout the whole of the visual pathway following visual loss, particularly in amblyopia. MR spectroscopy techniques have also revealed several abnormalities in metabolite concentrations in both glaucoma and age-related macular degeneration. GABA-edited MR spectroscopy on the other hand has identified possible evidence of plasticity within visual cortex.

Summary: Collectively, using MRI to investigate the effects on the visual pathway following disease and dysfunction has revealed a rich pattern of results allowing for better characterisation of disease. In the future MRI will likely play an important role in assessing the impact of eye disease on the visual pathway and how it progresses over time.

Keywords: albinism; amblyopia; anophthalmia; functional magnetic resonance imaging; glaucoma; macular degeneration; magnetic resonance imaging; magnetic resonance spectroscopy; ophthalmology; visual deficit.

Figure 1

Diffusion measurements reveal white matter tracts. (a) Coronal section through a typical diffusion weighted data volume; dark areas represent areas of relatively high diffusion in a single tested diffusion direction. (b) Diffusion measures are routinely taken in multiple directions. A ‘mean diffusivity’ can be computed by averaging over all diffusion directions as shown in (b); again dark areas represent high ‘mean diffusivity’. (c) This image reveals how directional, or anisotropic, the diffusion is at any given location; areas of high intensity have a high ‘fractional anisotropy’ (FA) and are thus biased for diffusion in a specific direction. (d) The false colour map overlaid on the FA image represents the preferred direction of diffusion at each location; blue colours represent preferred diffusion in the z‐plane (superior‐inferior), red in the x‐plane (left‐right) and green in the y plane (anterior‐posterior). (e) A zoomed representation of the preferred diffusion direction that can be represented by coloured unit line vectors at each location.

Figure 2

Normal MR spectrum showing the chemical shift in parts per million (ppm) for the metabolites myo‐Inositol (Ins), choline (Cho), creatine (Cr), GABA, Glutamate, Glutamine complex (Glx), N‐acetylaspartate (NAA) and lipids.

Figure 3

GABA‐edited spectrum produced from a MEGA‐PRESS sequence using the processing tool GANNET. The GABA‐edited data is represented by the blue line with the model of best fit represented in red. The black line below represents the residual between the model and the data.

Figure 4

Retinotopic mappings in individuals with eye disease. In each panel, an inflated rendering of the occipital lobe (a magnified version of the area highlighted by the broken line in the hemisphere shown to the left) of the left hemisphere is illustrated. Superimposed on the occipital lobes in false colour are the visual field locations that the brain represents. In each panel the colour coding of the visual field location is given above the visualisations of the brain maps. In (a) and (b), the map of polar angle and eccentricity in a control are given. Polar angle is used to determine boundaries between visual field maps. The eccentricity maps of this control participant are adapted in panels (c–e) to schematise the results observed in individuals with retinal scotomas. In panel (c) a schematic map of eccentricity is shown for a rod achromat alongside, in panel (d), a schematic of a control viewing under scotopic conditions. The broken circle highlights where differences in the map were found; the rod achromat has activity that extends more posteriorly than in the control indicating that the visual cortex has reorganised. In (e), the schematic eccentricity map for an individual with Age‐related Macular Degeneration, in which no evidence of large‐scale reorganisation is shown. In f and g, the eccentricity maps obtained for stimulation of the nasal (f) and temporal (g) retina of the right eye of an individual with albinism are given. Stimulation of the nasal retinal of the right eye produces a map that is very similar to that derived from controls. In contrast to controls however a large representation of the ipsilateral visual field is seen when the temporal retina of the right eye is stimulated. The maps show that a given cortical location responds to equal, but opposite eccentricities.