Multi-time scale dynamic effective brain networks reveal accelerated brain aging in individuals with major depressive disorder.

The integrated feature set combining all temporal scales (ALL) achieved the highest brain age prediction accuracy in healthy controls.

• The ALL feature set achieved a mean absolute error (MAE) of 3.64 years in healthy controls.
• Rs-fMRI data were collected from 80 healthy controls and 80 MDD patients.
• Time-series signals were extracted from 116 brain regions to construct dynamic effective networks across four temporal scales using a coarse-graining algorithm.
• A support vector regression model was trained using data from the HC group to estimate brain age.

The brain age gap (BAG) was significantly higher in current MDD patients compared to healthy controls.

• Mean BAG was 1.96 years for healthy controls and 4.56 years for current MDD patients.
• Post hoc tests with Bonferroni correction showed a significant difference between current MDD and HC (t = 4.85, p < 0.001).
• The current MDD subgroup included 46 participants.

The brain age gap (BAG) was significantly higher in remitted MDD patients compared to healthy controls.

• Mean BAG was 3.16 years for remitted MDD patients compared to 1.96 years for healthy controls.
• Post hoc tests with Bonferroni correction showed a significant difference between remitted MDD and HC (t = 2.72, p = 0.009).
• The remitted MDD subgroup included 34 participants.

There was no significant difference in brain age gap between current MDD and remitted MDD groups.

• Post hoc tests with Bonferroni correction found no statistically significant difference in BAG between current MDD (mean BAG = 4.56 years) and remitted MDD (mean BAG = 3.16 years).
• Both current and remitted MDD groups showed significantly elevated BAG relative to healthy controls.

No significant correlations were found between brain age gap and clinical measures of depression.

• BAG showed no significant correlation with depression duration.
• BAG showed no significant correlation with HAMD (Hamilton Depression Rating Scale) scores.
• This finding applied across the MDD groups examined.

Multi-time scale dynamic effective brain networks were constructed from resting-state fMRI across four temporal scales.

• A coarse-graining algorithm was used to generate four temporal scales from rs-fMRI time-series data.
• Time-series signals were extracted from 116 brain regions of interest.
• An integrated feature set (ALL) was created by combining features across the four temporal scales.
• The study included 80 healthy controls and 80 MDD patients (46 in current phase, 34 in remitted phase).