Functional interactions in patients with hemianopia: A graph theory-based connectivity study of resting fMRI signal

Abstract

The assessment of task-independent functional connectivity (FC) after a lesion causing hemianopia remains an uncovered topic and represents a crucial point to better understand the neural basis of blindsight (i.e. unconscious visually triggered behavior) and visual awareness. In this light, we evaluated functional connectivity (FC) in 10 hemianopic patients and 10 healthy controls in a resting state paradigm. The main aim of this study is twofold: first of all we focused on the description and assessment of density and intensity of functional connectivity and network topology with and without a lesion affecting the visual pathway, and then we extracted and statistically compared network metrics, focusing on functional segregation, integration and specialization. Moreover, a study of 3-cycle triangles with prominent connectivity was conducted to analyze functional segregation calculated as the area of each triangle created connecting three neighboring nodes. To achieve these purposes we applied a graph theory-based approach, starting from Pearson correlation coefficients extracted from pairs of regions of interest. In these analyses we focused on the FC extracted by the whole brain as well as by four resting state networks: The Visual (VN), Salience (SN), Attention (AN) and Default Mode Network (DMN), to assess brain functional reorganization following the injury. The results showed a general decrease in density and intensity of functional connections, that leads to a less compact structure characterized by decrease in functional integration, segregation and in the number of interconnected hubs in both the Visual Network and the whole brain, despite an increase in long-range inter-modules connections (occipito-frontal connections). Indeed, the VN was the most affected network, characterized by a decrease in intra- and inter-network connections and by a less compact topology, with less interconnected nodes. Surprisingly, we observed a higher functional integration in the DMN and in the AN regardless of the lesion extent, that may indicate a functional reorganization of the brain following the injury, trying to compensate for the general reduced connectivity. Finally we observed an increase in functional specialization (lower between-network connectivity) and in inter-networks functional segregation, which is reflected in a less compact network topology, highly organized in functional clusters. These descriptive findings provide new insight on the spontaneous brain activity in hemianopic patients by showing an alteration in the intrinsic architecture of a large-scale brain system that goes beyond the impairment of a single RSN.

Fig 1. Representation of each patients’ lesion.

Overlapped lesions of left (A) and right (B) damaged patients on the ch2.nii template of MRICron, represented on multislices. Each color represents one patient. (C). Quantification of the percentage of overlap between each patient’s lesion and the MNI Structural Atlas of fsl. TH: Thalamus; PUT = Putamen; INS = Insula; FRONT = Frontal Lobe; TEMP = Temporal Lobe; PAR = Parietal Lobe; OCC = Occipital Lobe.

Fig 2. Sensitivity analysis of networks’ density.

Sensitivity analysis showing changes in network density according to the absolute threshold applied to the pairwise correlation coefficients, within the range of r = 0.4–0.5, considering intra- and inter-network as well as the whole brain functional connectivity. VN = Visual Network; DMN = Default Mode Network; AN = Attentional Network; SN = Salience Network; RSNs = Resting State Networks.

Fig 3. Correlation coefficients higher in controls (A) or in the group of 10 patients (B).

Cortical surface representation of correlation coefficients extracted from the threshold (r>0.5) adjacent weighted undirected matrix, higher in controls than patients and vice versa, considering each RSN separately. A: Higher connections in controls than patients. B: Higher connections in the group of 10 patients than controls. VN = Visual Network; DMN = Default Mode Network; AN = Attentional Network; SN = Salience Network.

Fig 4. Correlation coefficients higher in controls (A) or in the group of 7 patients (B).

Cortical surface representation of correlation coefficients extracted from the threshold (r>0.5) adjacent weighted undirected matrix, higher in controls than patients and vice versa, considering each RSN separately. A: Higher connections in controls than patients. B: Higher connections in the group of 7 patients than controls. VN = Visual Network; DMN = Default Mode Network; AN = Attentional Network; SN = Salience Network.

Fig 5. “Spring” topology-based layout in different RSNs separately.

RSNs “spring” topology-based layout used to represent the correlation coefficients extracted from the adjacent unweighted undirected matrix (r>0.5), in the group of controls (HC), in the group of 10 patients (PT10) and in the group of 7 patients (PT7). SN = Salience Network; DMN = Default Mode Network; AN = Attentional Network; VN = Visual Network.

Fig 6. Matrices of correlation coefficients including regions belonging to all RSNs, showing intra- and inter-network correlations.

Adjacent weighted undirected correlation matrices (r>0.5) of controls, of 10 patients and of 7 patients, representing intra- and inter-network connectivity. A greater number of positive between-networks correlations was observed in the group of controls. SN (Salience Network): Superior Medial Frontal Cortex, Insula, Anterior, Middle and Posterior Cingulate. DMN (Default Mode Network): Hippocampus, Parahippocampal Gyrus, Fusiform Gyrus, Angular Gyrus, Precuneus, Middle Temporal Cortex. AN (Attentional Network): Middle Frontal Gyrus, Middle Frontal Gyrus Orbital Part, Inferior Parietal, Superior Temporal Gyrus, Superior Frontal Gyrus (dorsolateral), Inferior Frontal Gyrus (opercular, triangular and orbital part), Superior Parietal Gyrus. VN (Visual Network): Calcarine, Cuneus, Lingual Cortex, Superior, Middle and Inferior Occipital Cortex.

Fig 7. Correlation coefficients higher in controls (A) and in the group of 10 patients (B).

Cortical surface representation of the correlation coefficients extracted from the threshold (r>0.5) adjacent weighted matrix, higher in controls than patients and vice versa, considering all RSNs together (left) and the whole brain (right). A: Higher connections in controls. B: Higher connections in the group of 10 patients. Red nodes indicate the Salience Network; Yellow nodes indicate the Default Mode Network; Green nodes indicate the Attentional Network; Blue nodes indicate the Visual Network.

Fig 8. Correlation coefficients higher in controls (A) and in the group of 7 patients (B).

Cortical surface representation of the correlation coefficients extracted from the threshold (r>0.5) adjacent weighted matrix, higher in controls than patients and vice versa, considering all RSNs together (left) and the whole brain (right). A: Higher connections in controls. B: Higher connections in the group of 7 patients. Red nodes indicate the Salience Network; Yellow nodes indicate the Default Mode Network; Green nodes indicate the Attentional Network; Blue nodes indicate the Visual Network.

Fig 9. “Spring” topology-based layout in the four RSNs (upper) and in the Whole Brain (lower).

“Spring” topology-based layout used to represent the correlation coefficients extracted from the adjacent unweighted undirected matrix of the RSNs (A) and of the whole brain (B), in the group of controls (HC), in the group of 10 patients (PT10) and in the group of 7 patients (PT7). SN = Salience Network; DMN = Default Mode Network; AN = Attentional Network; VN = Visual Network.

Fig 10. 3-cycle triangles in the VN, AN, SN and in the whole brain.

Representation of 3-cycle triangles area distribution in the group of controls (A), of 10 patients (B) and of 7 patients (C), for the VN, AN, SN and the whole brain.