USING MULTI-SCALE ENTROPY ANALYSIS IN FMRI DATA TO EXAMINE THE COGNITIVE FUNCTION IN PATIENTS OF LATE-LIFE DEPRESSION AFTER ONE-YEAR FISH OIL SUPPLEMENT

Abstract Background Late-life depression (LLD) is a risk factor for dementia and cognitive decline. Fish oil supplement is efficacious for depressive symptoms with cognitive protective benefits. While there is limited strategy to deter cognitive decline in older adults, examining the cognitive protection in LLD by using fish oil can potentially provide a new promise. Aims & Objectives In addition to using double-blind randomized controlled trial to examine the clinical effects of fish oil in LLD, we also explore the neural mechanism by using pre-and post fMRI scans. Method A 52-week trial was conducted. Complexity analysis (multi-scale sample entropy) was employed to analyze rs-fMRI data. We also correlate the MRI changes with changes in cognitive function and inflammatory profiles. Results Limited by the pandemic, we only included 20 patients with 11 u nder fish oil and nine under placebo. Only marginal improvement in processing speed was found in fish oil group. We found decreased brain entropy in left posterior cingulate gyrus (PCG), multiple visual areas, the orbital part of the right middle frontal gyrus, and the left Rolandic operculum. Brain entropy changes in PCG in the fish oil group was associated with language function improvement and decrease in interleukin-6 levels. Discussion & Conclusion Our results confirm the anti-inflammatory effects of fish oil in LLD could be the underlying mechanism of cognitive protective effects. Also, fish oil may exert its function through modulation of brain entropy changes in key area in default-mode network, a network strategic both in depression and dementia.