Loading...

Investigating the Effects of DBS on Brain Connectivity by Causal Inference in Parkinson’s Disease

Ostad Mohammadi, Mohammad | 2019

1218 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 52246 (19)
  4. University: Sharif University of Technology
  5. Department: Computer Engineering
  6. Advisor(s): Rabiee, Hamid Reza
  7. Abstract:
  8. Parkinson’s disease (PD) is a progressive debilitating neurological disorder that causes motor and cognitive impairment. Administration of dopaminergic medication (Levodopa) has been reported to be effective in attenuating the excessive pathological synchronization in basal ganglia. However, long term levodopa therapy has its pitfalls. High frequency deep brain stimulation (DBS) has been suggested as an effective alternative for reducing motor symptoms in PD. In this method, distinct brain regions involved in the pathophysiology of the disease are stimulated electrically at high frequencies (i.e. at 130 Hz). While several studies have been carried out on the effects of DBS and its clinical efficiency, its precise mechanism still remains elusive. Investigating the effects of DBS on brain connectivity might lead us to a better understanding of DBS mechanism in order to optimize this method for better outcomes. By inferring functional connectivity within brain regions from multiple simultaneous time series, such as magnetoencephalography (MEG), an intuitive understanding of neurodegenerative disorder can be obtained. Recently, causal inference has been utilized to examine functional connectivity. Causal inference is a statistical method to determine which causal connections between nodes in a network can be concluded from empirical observations even in presence of latent variables. In this research, we aim to find the effects of DBS on the directed connectivity between different brain regions by using causal discovery in PD using the most popular method, i.e. Granger Causality.
    At this research, we have found some criterions that differ between DBS ON and OFF modes. These criterions can be devided into two groups of graph measures and statistical tests that investigate power of connections in the two states. DBS improves Parkinson patient’s motor performance, so these criterions can be used to improve diagnosis of Parkinson and the effects of DBS method
  9. Keywords:
  10. Parkinson Disease ; Deep Brain Stimulation (DBS) ; Causal Inference ; Granger Causality

 Digital Object List