Objective:

To examine resting-state functional differences between patients with post-stroke depression (PSD) and healthy controls, and to evaluate the use of machine learning to identify imaging features associated with PSD.

Approach:

Key Findings:

• PatientswithPSDshowedresting-statefunctionaldifferencesincingulate,thalamic,prefrontal,insular,posteriordefault-mode,andvisual-associatedregions.Twenty-ninecandidatefeaturesdifferedbetweengroups,including7ReHo,8ALFF,6DC,and8FCfeatures.LASSOretained10corefeatures,withacross-validatedAUCof0.878.TheExtraTreesmodelachievedthehighestindependenttest-setperformancewithanAUCof0.889.KeyinfluentialfeaturesincludedDCintheleftanteriorcingulate,ReHointheleftthalamus,andFCbetweentheleftprecuneusandleftcalcarinecortex.

Interpretation:

The study identified multi-level resting-state functional differences in PSD, with machine learning revealing candidate rs-fMRI features.

Limitations:

• Findings require validation in larger longitudinal cohorts.
• The specificity and clinical utility of identified features need further clarification.

Conclusion:

The study highlights the integration of machine learning with rs-fMRI to understand PSD-related brain function variations.

Attribution Notice

This content is an AI-generated, fully rewritten summary based on a published scholarly article. It does not reproduce the original text and is not a substitute for the original publication. Readers are encouraged to consult the source for full context, data, and methodology.