Grant: CIHR Fellowship

Accumulating evidence has led to the reconceptualization of mental illnesses as syndromes of 'miswired brain connectivity' during brain development. To transform such insights into clinically useful deliverables, however, we need a precise delineation of disease-specific properties. Unfortunately, the current 'symptom-based' diagnosis fails to explain commonly encountered overlaps in clinical presentations as well as their heterogeneity. The key to address this level of complexity may lie in the creation of an objective, data-driven classification system based on an underlying biology. In the proposed research program, I will develop a novel 'disease subtyping' framework based on the brain network derived from resting-state functional magnetic resonance imaging (MRI). I will conduct my research based on the community-referred 'Healthy Brain Network Biobank', an ongoing initiative that collects multimodal MRI and behavioral data from 10,000 youth who may have psychiatric symptoms. My study targets >680 children affected by mental illness including attention-deficit/hyperactivity, anxiety, conduct, and autism spectrum disorders. Two analyses will be performed: 1) connectome-based disease subtyping and 2) computational network simulation. In the first analysis, I will extract the latent pathological components from functional networks underlying developmental disorders and cluster them to find more homogeneous subgroups. Multimodal profiling will then comprehensively evaluate the differences in behavior and brain organization across the identified subgroups. In the second analysis, a computational network simulation will estimate hidden biophysical features related to neurodynamics in order to probe disease mechanisms. Data-driven disease subtyping will provide objective transdiagnostic boundaries that mirror underlying biological processes. The proposed framework holds promise for the development of effective targeted therapies in multiple developmental conditions.