PREDICTING OUTCOME IN CHILDREN WITH NEW-ONSET TICS USING NEUROIMAGING DATA

点击翻译按钮获取中文摘要

 DESCRIPTION (provided by applicant): Childhood onset neuropsychiatric disorders are often debilitating and are increasing in prevalence. Tourette Syndrome/Chronic Tic Disorder (TS/CTD), we propose, has a course that lends itself especially well to discovering optimal strategies for early diagnosis and prevention of developmental disorders. Common clinical practice and epidemiological data suggest that a large fraction (10-25%) of all children manifest motor and/or vocal tics at some age. Yet only 1-3% of all children have tics for the full year required for diagnosis of TS/CTD. The objective of the proposed research is to understand the structural and functional neuroimaging features of children with new-onset tics ("New Tics"), as the neurobiology of this population has not been investigated. Studying children with New Tics should provide the most leverage in understanding why tics remit in some children but not others. This proposal compares neuroimaging features of New Tics to TS/CTD and to controls, implements machine-learning tools to predict whether children with New Tics will remit or develop TS/CTD, and utilizes longitudinal scans to identify within-subject changes that occur as tics remit or persist. The proposed study will apply resting-state functional connectivity MRI (rs fcMRI) and structural MRI methods to investigate children with New Tics. Neuroimaging will be conducted on children with current, new- onset tics, and comparisons will be made to existing MRI data from children with diagnosed TS/CTD and controls. Follow-up evaluation (1 year after tic onset) of New Tics will allow us to identify which children's tics remitted completely and whic were actually in the earliest stages of TS/CTD, thus sorting this group into Remitted Tics and Converted TS/CTD subgroups. Based on the epidemiological data, we expect most of New Tics subjects to fall in the Remitted Tics subgroup. Longitudinal scans of the New Tics children will enable us to assess within-subject changes that occur with remittance or persistence of tics. This study will also apply machine learning tools, specifically Support Vector Machines (SVMs), to characterize features that distinguish Remitted Tics and Existing TS/CTD groups. Those features will then be used to predict whether tics in the New Tics group will remit (Remitted Tics) or persist (Converted TS/CTD) on an individual patient basis. Thus, we will be poised to identify differences, and potentially early predictors, of remittance and TS/CTD. This study will provide innovative, important data on a common clinical presentation: the child with New Tics. Completing our aims successfully will allow individual prediction of remission or progression, or at least will allow a New Tics sample to be enriched for high risk of developing TS/CTD, which would make prevention studies for TS/CTD feasible. We will also gain insight into the neurobiology of New Tics. Finally, this study is an essential first step towards a definitive longitudinal study that can improve diagnostic accuracy and settle questions of cause and effect. With successful completion of our aims, similar methods can be applied to other childhood onset neuropsychiatric disorders, setting the stage for early treatment or prevention of chronicity.