Efficacy Evaluation Of Compressed Sensing Based DCE-MRI Sequence Combined With APTw For Neoadjuvant Chemotherapy In Breast Cancer

Objective:（1）To investigate the effect of different Acceleration Factor（AF）on the image quality of Dynamic Contrast Enhanced MRI（DCE-MRI）by Compressed SENSE（CS）,and to optimize DCE-MRI.（2）To explore the value of DCE-MRI combined with Amide Proton transfer-weighted imaging（APTw）in evaluating the efficacy of neoadjuvant chemotherapy（NAC）for breast cancer.Materials and Methods:A total of 101 patients who underwent breast MRI examination in our hospital from October 2019 to October 2020 were included.Three patients were excluded from the group due to poor image quality（motion artifact,low signal-to-noise ratio）,and 98 patients with an average age of（43.98±15.21）years were included in the final group.Based on the enhanced T1 high resolution isotropic volume examination（e-THRIVE）,different acceleration techniques（SENSE and CS）were used and combined with different AF（SENSE AF=4;CS AF=4,5,6,7）DCE-MRI scan of breast was performed on the patients.The effective signal noise ratio（SNR_E）and effective contrast noise ratio（CNR_E）of different AF images were measured by two observers at an ISP（Philips,Intellispace Portall Version7）workstation,and the image quality was subjective evaluated by five-point method.Another 31 female patients who underwent breast MRI examination in our hospital from October 2020 to January 2021 were included.Inclusion criteria:（1）Pathological results of puncture were breast cancer and NAC treatment;（2）MR examination was performed before and 2 weeks after NAC treatment;（3）The pathological results after NAC were complete.Exclusion criteria:（1）Previous history of breast cancer radiotherapy and chemotherapy（n=6）;（2）Poor image quality（n=1）.Twenty-four patients were eventually enrolled,with an average age of（54.00±7.20）years.Pathological response（PCR）group and non-PCR group were divided into the patients according to the pathological results of the operation after NAC.The T1WI,T2WI,APT and optimized DCE-MRI sequence scans were performed on Philips Ingenia CX 3.0T MRI device.Two radiologists used the vascular Permeability analysis software on the ISP workstation to measure the quantitative parameters of DCE-MRI by double-blind method,including volume transfer constant(K^trans),rate constant(K_ep),extracellular blood vessels outside the clearance volume ratio（Ve）,APT and tumor size（diameter,d）,and calculate the NAC quantitative difference before and after 2 weeks NACΔK^trans,ΔK_ep,ΔVe andΔAPT namelyΔd value and maximum tumor diameter difference value.SPSS 21.0 statistical software was used for analysis,and intraclass correlation coefficient（ICC）test was performed on the measured data of DCE-MRI images.ICC<0.4 indicates a consistency difference of 0.4-0.75,and ICC>of 0.75 indicates a good consistency.The consistency of subjective score data was tested by Kappa test.If the consistency is good,the subjective score and measurement data of senior physicians are selected for subsequent statistical analysis.Quantitative data were compared between groups using Kruskal-Wallis test,and the data with differences were selected for pairwise comparison.Chi-square test was used to compare differences in subjective scoring data.Quantitative data between the two groups were analyzed using two independent samples Mann-Whitney rank test.P<0.05 was considered statistically significant.Spearman was used to analyze the correlation between DCE-MRI quantitative parameters and APT value.Logistic regression analysis was used for data with differences between the two groups,and receiver operating characteristic curve（ROC）was used to analyze the diagnostic efficacy of DCE-MRI and APTw.Results:The image quality data obtained from DCE-MRI with different AF by two observers were in good agreement（ICC value was>0.75 for both）.The difference between SNR_{E lesions} and CNR_{E lesions} measured by DCE-MRI with different AF was statistically significant（P<0.05）,while there was no statistically significant difference between SNR_{E glands} with different AF（P=0.449）.SNR_{E lesions}and CNR_{E lesions} decreased significantly when AF=6 and 7（P<0.05）,and ubjective score decreased significantly when AF=5（P=0.048）.The optimized quantitative parameters of DCE between PCR and non-PCR groups were significantly different in Ktransbefore（P=0.003）,and statistically significant differences were found in K_after^trans,K_{ep after},Ve _after and APT_after（P<0.05）.The values of K^trans,K_ep,APT and d in PCR group were significantly different before and after NAC（P<0.05）.There were no significant differences in DCE parameters,APT values and d values before and after NAC in non-PCR group（P>0.05）.The p CR groupΔK^trans,ΔAPT andΔd is significantly higher than the p CR group（P<0.05）.ΔK_epΔVe and there was no statistically significant difference between the two groups（P=0.343,0.280）.ΔK_epwithΔAPT medium correlation（r=0.654,P<0.05）,ΔK^trans,Δthe V_ep andΔAPT no correlation（P=0.216,0.277）.ΔK^trans,Δd has good diagnostic performance（AUC=0.824,0.815）,the sensitivity of 100%and 66.7%respectively,specificity of 61.1%and 94.4%respectively.Conclusion:CS technique combined with different AF to achieve breast DCE-MRI has a good feasibility,and choosing the AF of CS as 4 can ensure the image quality unchanged while shortening the scanning time,which is shortened by10.8%.The optimized DCE-MRI sequence combined with APTw is helpful to predict and evaluate the efficacy of NAC in breast cancer patients.