Bi-threshold frequent subgraph mining for Alzheimer disease risk assessment

Fei Gao; Jing Li; Teresa Wu; Kewei Chen; Xiaonan Liu; Leslie Baxter; Richard J. Caselli

doi:10.1117/12.2293773

6 March 2018 Bi-threshold frequent subgraph mining for Alzheimer disease risk assessment

Fei Gao, Jing Li, Teresa Wu, Kewei Chen, Xiaonan Liu, Leslie Baxter, Richard J. Caselli

Proceedings Volume 10579, Medical Imaging 2018: Imaging Informatics for Healthcare, Research, and Applications; 105790C (2018) https://doi.org/10.1117/12.2293773
Event: SPIE Medical Imaging, 2018, Houston, Texas, United States

Abstract

An emerging trend in AD research is brain network development including graphic metrics and graph mining techniques. To construct a brain structural network, Diffusion Tensor Imaging (DTI) in conjunction with T1 weighted Magnetic Resonance Imaging (MRI) can be used to isolate brain regions as nodes, white matter tracts as the edge, and the density of the tracts as the weight to the edge. To study such network, its sub-network is often obtained by excluding unrelated nodes or edges. Existing research has heavily relied on domain knowledge or single-thresholding individual subject based network metrics to identify the sub network. In this research, we develop a bi-threshold frequent subgraph mining method (BT-FSG) to automatically filter out less important edges in responding to the clinical questions. Using this method, we are able to discover a subgraph of human brain network that can significantly reveal the difference between cognitively unimpaired APOE-4 carriers and noncarriers based on the correlations between the age vs. network local metric and age vs. network or global metric. This can potentially become a brain network marker for evaluating the AD risks for preclinical individuals.

Conference Presentation

Citation Download Citation

Fei Gao, Jing Li, Teresa Wu, Kewei Chen, Xiaonan Liu, Leslie Baxter, and Richard J. Caselli "Bi-threshold frequent subgraph mining for Alzheimer disease risk assessment", Proc. SPIE 10579, Medical Imaging 2018: Imaging Informatics for Healthcare, Research, and Applications, 105790C (6 March 2018); https://doi.org/10.1117/12.2293773

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available