paperKB
coga / coga-kb
Help
Sign in

Chunk #21 — Results — Discriminative features — EEG-FC

Source
Predicting alcohol use disorder remission: a longitudinal multimodal multi-featured machine learning approach.
Embedded
yes

Text

Default Mode Network (DMN) FC, as well as, connectivity levels in other brain networks, were found to discriminate between the groups and to predict AUD remission (selected significance FC features according to groups in Table 3 and Fig. 2. Full list of FC features including right, and left hemispheres can be found in Supplementary Tables S11–S15). Lower connectivity of DMN ROIs was found in the continued AUD group, especially in the range of theta and gamma bands. Known DMN hubs including the precuneus (AA female), the posterior cingulate (AA female), and the middle frontal (EA male, AA male) showed lower gamma connectivity. Lower Theta connectivity was found in other more temporal DMN hubs such as the temporal cortex (AA female) and the parahippocampal formation (EA female) with anterior and posterior brain areas. Lower theta connectivity in the continued AUD versus remitted group was also found between the fusiform and posterior and anterior brain areas. Higher connectivity in the continued AUD group was found in selected temporal and parietal areas including the insula connectivity with superior parietal (beta, EA female) and with inferior parietal (theta, AA male) (Fig. 2).