RT Journal Article
SR Electronic
T1 Categorizing SHR and WKY rats by chi2 algorithm and decision tree
JF bioRxiv
FD Cold Spring Harbor Laboratory
SP 2020.02.03.931899
DO 10.1101/2020.02.03.931899
A1 Ping-Rui Tsai
A1 Kun-Huang Chen
A1 Tzay-Ming Hong
A1 Fu-Nien Wang
A1 Teng-Yi Huang
YR 2020
UL http://biorxiv.org/content/early/2020/02/03/2020.02.03.931899.abstract
AB In the past two decades neuroscience has offered many popular methods for the analysis of mental disorder, such as seed-based analysis, ICA, and graph methods. They are widely used in the study of brain network. We offer a new procedure that can simplify the analysis and has a high ROC index over 0.9. This method uses the graph theory to build a connectivity network, which is characterized by degrees and measures the number of effective links for each voxel. When the degree is ranked from low to high, the network equation can be fit by the power-law distribution. It has been proposed that distinct and yet robust exponents of the power law can differentiate human behavior. Using the mentally disordered SHR and WKY rats as samples, we employ chi2 algorithm and Decision Tree to classify different states of mental disorder by analyzing different traits in degree of connectivity.