Objective: To build a frame of whole brain anatomical network for incorporating different data from different studies with different brain parcellations and resolutions, to evaluate the role of each brain region in advanced Parkinson’s disease (PD) brain in an approach of complex network analysis, and to apply the framework in modeling theories or hypotheses related to PD’s pathomechanism.

Background: PD is the most common neurodegenerative movement disorder and there is no cure so far. Understanding the roles and their changes of brain regions from normal to pathophysiological condition in a view of system and network is prospective and also important for improving the treatment protocol or finding new targets for both medicine and deep brain stimulation. Although numerous functional studies reported a few changes of region of interest by their non-invasive imaging methods, few studies have reported on the whole anatomical network.

Methods: Our study first proposed a series of algorithms to build a versatile frame for whole anatomical neural network based on a rhesus macaque database. Then we simulated PD with substantia nigra removal and an interdisciplinary complex network analysis was made to understand the changes taking place with PD. The huge detailed results were summarized with dimensionality reduction method (eg. principal component analysis). Lastly, the anatomical neural network was use to modeling some of the PD’s theories and hypotheses.

Results: The dimensionality reduced results revealed that the areas including the basal ganglion (striatum and globus pallidus), limbic system (amygdala), cortex (prefrontal lobe, visual cortex, insula) thalamus, hippocampus, etc., showed relatively notable drifts in their own patterns. Newer theory including Braak stage hypothesis and propagation of pathological protein can be simulated in our frame.

Conclusions: By using our approach, one can easily establish and use the whole anatomical neural network model and conduct the analysis as needed. Our model frame with enough stability and capacity can support to incorporate multimodal data (pathological, imaging, fMRI, EEG and so on), thus further studies on it should be proceed.

« Back to 2016 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/a-whole-brain-anatomical-network-model-and-its-application-in-pd-research/