Network-based biOmarker Discovery of Neurodegenerative Diseases Using Multimodal Connectivity

Multimodal connectivity graph metrics across diseases

Multi-layer graph combining the functional and structural connectivity will provide an ideal framework for identifying multimodal features. The first layer corresponds to the functional connectivity where links represent the similarity between the fMRI signal from different cerebral regions (i.e. the nodes), using the cross-correlation. The second layer represents the structural connectivity where the edges are the fiber density between the nodes. Topological metrics of graph: 1/ the nodal centrality which quantifies how important a node is within a network, 2/ the betweenness defined as the ratio of the number of the shortest paths comprising the node to the total number of shortest paths in the graph, measures the hub property of the node and 3/ local efficiency which quantify the ability of a network to transmit information at the global and local level, will be compared between patients at-risk for Alzheimer's Disease, patients with Parkinson's Disease and healthy controls.

Multimodal connectivity graph metrics across stages of disease

Multi-layer graph combining the functional and structural connectivity will provide an ideal framework for identifying multimodal features. The first layer corresponds to the functional connectivity where links represent the correlation between the fMRI signal. The second layer represents the structural connectivity where the edges are the fiber density between the nodes. Topological metrics of graph: 1/ the nodal centrality which quantifies how important a node is within a network, 2/ the betweenness defined as the ratio of the number of the shortest paths comprising the node to the total number and 3/ local efficiency which quantify the ability of a network to transmit information, will be compared between patients at-risk for Alzheimer's Disease at the Subjective Cognitive Decline stage and at the Mild Cognitive Impairment stage. Similarly, patients with Parkinson's Disease without cognitive impairment and with Parkinson's Disease with Mild Cognitive will be compared.

Correlation between multimodal connectivity graph metrics and cognitive scores

Cognitive scores will include standard tasks (MOCA, Symbol Digit Modality Test - oral, Digit span, 16-items Free \& Cued Selective Reminding Test, 10/36 spatial recall test, Oral Letter-Number Sequencing test, D-KEFS Color-Word Interference Test, Benton Judgment of Line Orientation, CLOX clock-drawing test, Boston Naming Test, Animal fluency) and an experimental recognition memory task. As for the latter, the scores used will be the congruency effect during study (RT in milliseconds), global recognition accuracy, and the effect of congruency on subsequent memory (difference in recognition accuracy between congruent and incongruent trials at study).