Individual Metabolic Neural Network Analysis in Pet Images for Differentiation Diagnosis and Prediction in Alzheimer Diseases

FDG PET imaging is a powerful imaging tool for investigating brain metabolism, but some difficulties exist for the identification of Alzheimer's disease (AD) and mild cognitive impairment (MCI). In recent years, graph theory has been applied to build a network of brain structure and function and provided an alternative method for studying the brain function. However, previous metabolic connectivity methods are based on groupwise data analysis for FDG metabolic network. The goal of this project is to derive an individual metabolic network used the FDG imaging for the first year grant period, and analysis of global and local network link situation as the differential diagnosis of AD and MCI for the second year grant period, and for the third year grant period, we will build up a disease progression prediction model, as well as a deep learning based diagnosis model for dementia using the individual connectivity brain network. Finally, this project is expected to be able to assist with early diagnosis of AD and MCI, as well as to assess converging for the time MCI into AD conversion rate of the year. The result can be as an effective tool to assess and reduce the cost of AD and MCI diagnosis, improve the quality of life of patients and can be developed as a research platform for the treatment development in Alzheimer's disease.

Project ID:PC10907-1549
External Project ID:MOST109-2314-B182-019-MY3