Objective To analyze the performance of amplitude of low frequency fluctuation （ALFF） in predicting the therapeutic efficacy of escitalopram in patients with depressive dissorder based on functional magnetic resonance imaging （fMRI） technology. Methods From April 2017 to June 2021， 87 patients with depressive disorder who visited the Outpatient Department of Beijing Anding Hospital， Capital Medical University were selected as the research subject. All patients received 12 weeks of treatment with escitalopram. The efficacy was evaluated using the Hamilton Depression Scale-17 items （HAMD-17） at baseline and after 12 weeks of treatment. Patients with a post-treatment score of ≤ 7 points were included in the remission group （n=42）， while those with a score greater than 7 points were included in the non-remission group （n=45）. During the baseline period， fMRI scans were performed on two groups of patients and ALFF values were calculated. The brain regions with differences in ALFF values between the two groups were compared. Pearson correlation was used to analyze the association between ALFF values in different brain regions and HAMD-17 scores. Receiver operating characteristic （ROC） curve was used to test the performance of differential brain region ALFF values in predicting the therapeutic efficacy of escitalopram. Results The baseline ALFF values of the bilateral calcarine gyrus/cuneiform gyrus and right lingual gyrus/parahippocampal gyrus in the non-remission group were higher than those in the remission group， and the difference was statistically significant （PFWE＜ 0.05）. The ALFF values of different brain regions were positively correlated with the HAMD-17 score of patients after 12 weeks of treatment （P＜ 0.05）， and negatively correlated with the HAMD-17 score reduction rate （P＜ 0.05）， and the differences were statistically significant. ROC curve analysis showed that the comprehensive prediction of the treatment efficacy of escitalopram was best based on the ALFF values of two different brain regions in the bilateral calcarine gyrus/cuneiform gyrus and right lingual gyrus/parahippocampal gyrus， with an area under the curve of 82.6%， sensitivity of 60.0%， and specificity of 95.2%. Conclusions The baseline local brain functional activity characteristic of depressive dissorder patients can serve as the potential marker to predict the efficacy of antidepressants， while enhanced functional activity in the visual network brain region may be a predictive factor for the poor therapeutic efficacy of escitalopram.