BACKGROUND: Functional magnetic resonance imaging (fMRI) is a non-invasive technique providing valuable insights into brain activity. Its application in stroke rehabilitation is emerging as a potential tool for predicting recovery outcomes. OBJECTIVE: This study aimed to explore the predictive capacity of baseline fMRI brain activation patterns on rehabilitation outcomes in stroke patients. METHODS: With 150 stroke patients, we conducted a prospective cohort research to examine the correlation between baseline fMRI results and 6-month rehabilitation outcomes. The modified Rankin Scale (mRS) and the National Institutes of Health Stroke Scale (NIHSS) were used as outcome indicators. RESULTS: Our findings revealed that the degree of brain activation within motor-related regions at baseline significantly correlated with functional improvement after six months. Both NIHSS score (R^2 = 0.45, p<0.001) and mRS score (R^2 = 0.38, p<0.001) were strongly associated with baseline fMRI data. CONCLUSION: Our results suggest that baseline fMRI brain activation patterns can predict rehabilitation outcomes in stroke patients, providing a promising direction for individualizing stroke rehabilitation strategies.