Altered spatio-temporal state patterns for functional dynamics estimation in first-episode drug-naive major...

Area of Science:

• Neuroscience
• Psychiatry
• Medical Imaging

Background:

• Major depressive disorder (MDD) is characterized by affective and cognitive deficits.
• While brain structure and function abnormalities in MDD are studied, large-scale spatio-temporal network alterations remain less understood.

Purpose of the Study:

• To investigate spatio-temporal functional network abnormalities in first-episode drug-naive (FEDN) MDD patients using a large dataset.
• To develop a classification method to identify discriminative brain states associated with FEDN MDD.

Main Methods:

• Utilized data from the International Big-Data Center for Depression Research (IBCDR).
• Employed multivariate pattern classification and independent component analysis to extract and analyze spatio-temporal brain states.
• Developed a multi-kernel pattern classification method to identify discriminative states in 314 FEDN MDD patients and 543 healthy controls (HC).

Main Results:

• FEDN MDD patients exhibited altered spatio-temporal functional states in the default mode network (DMN), salience network, a hub network, and a complex coupling network.
• Multi-kernel classification models achieved an area under the receiver operating characteristic curve of up to 0.80 for distinguishing patients from HC.
• Classification scores correlated with Hamilton Depression Rating Scale scores and age at MDD onset.

Conclusions:

• First-episode drug-naive MDD patients display multiple abnormal spatio-temporal functional brain states.
• The classification scores derived from these states are related to symptom severity.
• Assessing spatio-temporal states shows potential as a clinical and research tool for differentiating neuropsychiatric patients.