Objective To explore the clinical application of magnetic resonance imaging neoadjuvant chemotherapy of locally advanced breast cancer, and analyze the feasibility of the method as a predictor of response to chemotherapy. Methods 12 cases of patients with breast cancer after neoadjuvant chemotherapy on tumor volume change into effect group and ineffective group, the rate of change in tumor volume loss, maximum signal comparison before and after chemotherapy and chemotherapy before the maximum signal loss rate from a correlation between the rate and tumor. Results The effect of chemotherapy group, the average tumor volume of (2.89±1.03) cm3, significantly lower than that before chemotherapy and after chemotherapy the tumor volume of invalid should group, the difference was statistically significant (P< 0.05); group effect in patients receiving neoadjuvant chemotherapy, the maximum signal loss rate was (13±11.2)%, significantly below before chemotherapy and after chemotherapy should be related to invalid index group. Neoadjuvant chemotherapy after a period of treatment, loss rate change perfusion maximum signal between two groups was statistically significant (P> 0.05). There was a positive correlation between the shrinking rate of mean values of maximum signal loss rate and tumor volume (r=0.313, P=0.44). Conclusion Neoadjuvant chemotherapy of locally advanced breast cancer can be early evaluation of the application of magnetic resonance imaging technology, but need further study to prove whether it can be used as a predictor of response to chemotherapy.