2.12.2020 What group analysis brings (Wang)

With the advances of sensor technologies, we can access multiple datasets that are complementary or linked together. Generally, the phenomena that appear in most datasets are more attractive than those in the specific dataset. This dissertation aims to study how to extract interesting activities from brain imaging data at the group level.
M.Eng. Xiulin Wang
Published
2.12.2020

"Most conventional analysis methods in brain data cannot make full use of the inherent structural information and prior coupling information within/among datasets. Undoubtedly, in the era of big data, the high computational cost of processing large-scale datasets is also one of the existing issues. Therefore, we need more efficient and robust analytical methods for the emerging huge amount of brain data", suggests M.Eng. Xiulin Wang in his dissertation.

Xiulin Wang's dissertation introduces a series of advanced coupled matrix/tensor-based methods. These methods have been applied to the joint analysis of brain imaging data to discover the commonly stimulus-elicited features among subjects and extract the expected multi-domain features that can discriminate different groups.

"The proposed methods can be regarded as a generalization of matrix/tensor factorization to multiple datasets, providing a natural framework for the simultaneous extraction of shared and individual information. More importantly, they can significantly improve the computation efficiency without losing the decomposition accuracy and stability", Xiulin tells.

Xiulin Wang's research is currently focused on the group analysis of multi-subject and single-modal brain data, and has shed new light on how to analyze multi-subject and multi-modal brain data in cognitive neuroscience.

M.Eng. Xiulin Wang defends his doctoral dissertation in Software and Communications Engineering "Coupled Nonnegative Matrix/Tensor Factorization in Brain Imaging Data". Wednesday 2.12.2020 starting at 11. Opponent Dr. Qibin Zhao (Team Leader, RIKEN Center for Advanced Intelligence Project, Japan) and Custos Senior Reseacher Zheng Chang (Ä¢¹½Ö±²¥). The doctoral dissertation is held in English.

The audience can follow the dissertation online.
Link to the Zoom Webinar (Zoom application or Google Chrome web browser recommended):
Phone number to which the audience can present possible additional questions at the end of the event (to the custos): +358 40 8054143