Neural Correlates Associated to Images for Emtional Response

Course Project for Fundamentals of MRI and Neuroimaging Analysis

project status: completed (september - december 2024) :link: Project Report

This project maps how the brain processes emotionally positive, negative, and neutral images using a multimodal fMRI workflow on a 3T Siemens Prisma, combining structural MRI, resting-state fMRI, task-based BOLD, and ASL perfusion to link anatomy, connectivity, and blood flow with stimulus-driven responses. The functional task presented 100 images (40 positive, 40 negative, 20 neutral) with 8-second events and GLM analysis using a double-gamma HRF, alongside structural brain extraction/segmentation (T1/T2), ICA of resting-state data, and pCASL-based perfusion quantification.

results from the independent component analysis
general linear model

Key outcomes were clean structural segmentations supporting downstream alignment, GLM contrasts referencing amygdala and ventral striatum masks, and perfusion estimates of 101.47 ml/100g/min (grey matter), 80.78 ml/100g/min (white matter; elevated likely from partial volume), and 118.06 ml/100g/min (CSF; likely noise contamination), demonstrating a full pipeline from acquisition to quantitative interpretation of emotion-related brain function.