Loading...

Estimation of Brain Connectivity Via Deep Neural Network

Khodabakhsh, Alireza | 2022

65 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 56122 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Shamsollahi, Mohammad Bagher
  7. Abstract:
  8. The human brain is one of the most complex and least understood systems in nature. In recent decades, numerous studies have been conducted to identify the behavior of this system. One of the areas of brain research is the investigation of the connections between different regions of the brain during a presumed process or in a resting state. Among various types of brain connections, effective connectivity provides researchers with higher-level information on brain behavior compared to other connections, but also entails greater computational complexity. In recent years, researchers have aimed to provide an estimator with the maximum desirable capabilities, and with the advent of (deep) neural networks, achieving such a goal has become smoother. However, a considerable number of these estimators have been based on non-interpretable networks that infer causal relationships through supervised learning. In this study, based on the concept of Granger causality, we present an interpretable network called nCI-GAN, which infers causal relationships through unsupervised learning based on adversarial generative networks. This interpretable network can independently learn the order of each channel. Furthermore, nCI-GAN can infer causal relationships with fewer parameters compared to other (deep) neural network estimators in the face of high-dimensional time series data. Additionally, nCI-GAN exhibits stability in similarity metrics with a noticeable error compared to other GAN-based estimators
  9. Keywords:
  10. Granger Causality ; Time Series Analysis ; Generative Adversarial Networks ; Unsupervised Learning ; Interpretable Models ; Deep Neural Networks ; Brain Connectivity

 Digital Object List

 Bookmark

...see more