A biomarker discovery framework for childhood anxiety

Introduction Anxiety is the most common manifestation of psychopathology in youth, negatively affecting academic, social, and adaptive functioning increasing risk for mental health problems into adulthood. disorders are diagnosed only after clinical symptoms emerge, potentially missing opportunities to intervene during critical early prodromal periods. In this study, we used a new empirical approach extracting nonlinear features electroencephalogram (EEG), with goal discovering differences brain electrodynamics that distinguish children anxiety from healthy children. Additionally, examined whether could externalizing anxiety. Methods We novel supervised tensor factorization method extract latent factors repeated multifrequency EEG measures longitudinal sample assessed infancy at ages 3, 5, 7 years age. first validity by showing calendar age highly correlated complexity ( r = 0.77). then computed separately distinguishing controls using 5-fold cross validation scheme similarly controls. Results found derived recordings were required an disorder controls; infancy, 3 years, or 5 alone insufficient. However, two (5, years) three (3, gave much better results than year alone. Externalizing be detected 3- data, also giving any single snapshot. Further, sex assigned birth was important covariate improved accuracy both groups, birthweight as modestly disorders. Recordings infant did not contribute classification either Conclusion This study suggests extracted childhood promising candidate biomarkers if chosen appropriate ages.