Graph theoretical analysis of visual system functional connectivity in normal and lesioned brains
MetadataShow full item record
The visual system consists of two distinct networks--the structural and functional networks. Defined anatomically by the physical connections between neurons, the structural network provides the architectural framework for information processing. The functional network, however, is determined by the situation-dependent activation of connections either set by or independent of the structural network. Therefore, these two networks are linked. Individually, these networks can be mapped and analyzed using graph theory--a specialized branch of mathematics that is used to determine characteristics of complex networks. In the case of the visual system, nodes (objects of interest) are represented by specific brain regions of the visual network, and edges represent the structural (anatomical) or functional (physiological) connections between visual regions. The objective of this study was to use graph theory to determine the extent to which the structural and functional networks of the cat visual system are linked and to use this information to generate a hierarchy of visual information flow and processing through the visual system. Using five intact and six lesioned subjects, we analyzed sixteen visual cortical areas as well as nine sub-cortical areas and their connections. These regions include cortical areas A17, A18, A19, A20a, A20b, A21a, A21b, 7, PMLS, PLLS, AMLS, ALLS, DLS, VLS, PS, and SVA and sub-cortical areas LGN, LPL, LPM, PUL, CAUD, SGS, SO, SGI, and SGP. Structural connectivity data was obtained from literary sources. Functional connectivity data was obtained from 2-Deoxy-D-Glucose stained images of the subjects' brains. We first looked at the functional networks of the intact subjects to determine the functional influence of one hemisphere on the other and to observe general trends in network connections. We found that functional connections across hemispheres were not symmetrical and that two distinct connectional patterns emerged - visual areas, or clusters, may connect to their equivalent areas across hemispheres or they may connect to a many non-equivalent areas. There also emerged a distinct clustering pattern across and within both hemispheres. Clusters were seen within the primary visual cortical areas (17, 18) and area 19; between areas PMLS, PLLS, DLS, and VLS; and between the superior colliculi and all of these regions. We also used this data to determine which functional and structural connections are linked in resting state visual network and how strongly they are linked. We found that areas 17, 18, 19, 21a, PMLS, PLLS, DLS, and VLS are the ones primarily involved in visual processing in the passive state. The second group we looked at consisted of the subjects with a lesioned hemisphere either receiving sham or real trans-cranial magnetic stimulation. This group allowed us to determine the validity of predictions concerning the existence and influence of intrinsic inhibition between hemispheres. We found that when one hemisphere is lesioned, inhibition is released and there is an increase in functional connectivity within the remaining hemisphere. When TMS was performed and inhibition reinstated, this increase in functional connectivity was lost. This confirmed that there is an inhibitory influence of the visual system between hemispheres in the cat--meaning that in intact brains, one hemisphere decreases the functional activity of the other. These finding confirmed that, functionally, there are both inter- and intra-hemispheric influences between hemispheres and that altering part of the network will subsequently alter the functional organization and activity of the whole network.
Thesis (M.A.)--Boston University PLEASE NOTE: Boston University Libraries did not receive an Authorization To Manage form for this thesis or dissertation. It is therefore not openly accessible, though it may be available by request. If you are the author or principal advisor of this work and would like to request open access for it, please contact us at email@example.com. Thank you.