Fig. 7: Susceptibility connectivity as features for identifying individuals with cognitive impairment. | Translational Psychiatry

Fig. 7: Susceptibility connectivity as features for identifying individuals with cognitive impairment.

From: The characteristic patterns of individual brain susceptibility networks underlie Alzheimer’s disease and white matter hyperintensity-related cognitive impairment

Fig. 7

A For the HC vs MCI-AD classification, 263 edges as the features selected by the SLR are primarily distributed throughout the DMN-related and VN-related connectivity. The cross validation accuracy and AUC were 76.12% and 0.83, with a sensitivity and specificity of 76.32% and 75.86%, respectively. B For the WMH-NC vs WMH-MCI classification, 199 edges as the features selected by the SLR are primarily distributed throughout the SN-related and DMN-related connectivity. The cross validation accuracy and AUC were 84.85% and 0.93, with a sensitivity and specificity of 87.10% and 82.86%, respectively. HC health control, NC normal cognition, MCI-AD mild cognitive impairment due to Alzheimer’s disease, WMH white matter hyperintensities, VN visual network, SMN somatomotor network, DAN dorsal attention network, SVAN salience/ventral attention network, LN limbic network, FPN frontoparietal network, DMN default mode network, SN subcortical network, SLR sparse logistic regression, AUC area under curve.

Back to article page