Abstract
Machine learning classification techniques are frequently applied to structural and resting-state fMRI data to identify brain-based biomarkers for developmental disorders. However, task-related fMRI has rarely been used as a diagnostic tool. Here, we used structural MRI, resting-state connectivity and task-based fMRI data to detect congenital amusia, a pitch-specific developmental disorder. All approaches discriminated amusics from controls in meaningful brain networks at similar levels of accuracy. Interestingly, the classifier outcome was specific to deficit-related neural circuits, as the group classification failed for fMRI data acquired during a verbal task for which amusics were unimpaired. Most importantly, classifier outputs of task-related fMRI data predicted individual behavioral performance on an independent pitch-based task, while this relationship was not observed for structural or resting-state data. These results suggest that task-related imaging data can potentially be used as a powerful diagnostic tool to identify developmental disorders as they allow for the prediction of symptom severity.
Highlights
One of the main challenges of brain imaging is to provide individual discrimination ability to inform diagnosis and prognosis of neurodegenerative or developmental disorders at the individual level (Uddin et al, 2017)
Note that only a subset of subjects has participated to the rs-fMRI (13 in each group, all participants were recorded in Lyon) and task fMRI: pitch localizer (n = 12 in each group, all participants were recorded in Lyon)
Whole-brain searchlight analyses (Support Vector Machine, leave-one-out cross-validation procedure, permutation statistics, and cluster-level corrections) were performed on five different datasets consisting in: (A) a set of gray and white matter concentrations maps extracted from T1-MPRAGE volumes; (B) whole-brain rs-fMRI seed-based connectivity maps with seeds in the right and left auditory cortices; and (C) three task-fMRI datasets consisting of: a pitch localizer (Figure 1A), a short-term memory task for pitch (Figures 1B–D), and a short-term memory task for words
Summary
One of the main challenges of brain imaging is to provide individual discrimination ability to inform diagnosis and prognosis of neurodegenerative or developmental disorders at the individual level (Uddin et al, 2017). A growing number of studies have used machine learning classification techniques on either structural MRI (sMRI) or resting-state fMRI (rs-fMRI) data to identify brainbased disorder-related biomarkers (Arbabshirani et al, 2017; Uddin et al, 2017). These methods have shown great potential in discriminating abnormal development, such as Autism Spectrum Disorder, attention-deficit hyperactivity disorder or dyslexia, from typical development (Bray et al, 2009; Arbabshirani et al, 2017). In contrast, we hypothesized that task-based fMRI may present significant advantages in relating classifier outcomes to phenotypic or behavioral measures as compared to sMRI and rs-fMRI data because of the potential specificity they offer to probe brain activity
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
