Dynamic Evolution Mapping of Functional Brain Networks in De…

Research approach: Utilize high-time-resolution fMRI data to extract brain state transition sequences from hidden Markov models (HMM); construct a time-series convolutional network (TCN) to decode state evolution patterns and quantify the severity of depression and cognitive function impairment.

Innovation: Transforming brain network research from “static connections” to “dynamic evolution,” capturing millisecond-level abnormalities in brain state switching through sequential deep learning models, and revealing the essential characteristics of depression as a disorder of the brain’s dynamic system.

This manuscript is a secondary analysis study using open-access neuroimaging datasets.

The writer is expected to strictly follow neuroimaging journal standards (IMRAD structure), clearly describe data sources and preprocessing pipelines, ensure methodological reproducibility, and avoid any fabricated data or results.

The focus should be on individual-level functional organization and cross-diagnostic comparison.