Title:	Neuroimage analysis for Alzheimer's disease using deep learning
DNr:	NAISS 2023/22-509
Project Type:	NAISS Small Compute
Principal Investigator:	Hang Zhao <hang.zhao@physics.gu.se>
Affiliation:	Göteborgs universitet
Duration:	2023-05-23 – 2024-06-01
Classification:	30105
Keywords:

Abstract

The project aims to understand the mechanism of neurodegenerative diseases (ND) in the brain, especially with machine learning methods. In the last ten years, research has shown the brain connectivity pattern is a new biomarker of ND(1). Therefore, comprehensively grasping the brain connectivity change behind ND would be beneficial to the diagnosis and treatment of such diseases. Neuroimaging, as a stable, advanced imaging technique has promoted ND understanding. As the demand for neuroimage analysis rises, machine learning, especially deep learning, as a robust and efficient data analysis tool, is commonly applied to neuroimage analysis tasks to identify any possible biomarkers from brain connectivity for ND. In this project we will use the advances in neuroimaging and develop new deep learning image processing tools, as the project involves intensive neuroimaging data analysis, preprocessing of the imaging data is essential, thus, the decent CPU resource is appreciated also. 1. DelEtoile, J., & Adeli, H. (2017). Graph theory and brain connectivity in Alzheimer’s disease. The Neuroscientist, 23(6), 616-626.