摘要研究MRI纹理分析在预测新辅助化疗乳腺癌患者疗效中的应用价值。统计126例乳腺癌患者新辅助化疗（NAC）2周后疗效，并分析NAC不同疗效患者治疗前后MRI纹理参数（包括逆差矩、熵、能量及相关性）差异及其在鉴定乳腺癌NAC疗效中的价值。研究发现，治疗后熵、能量及相关性参数在预测NAC治疗疗效中均具有较好的效能（AUC=0.769，0.753，0.700，均 P<0.001），治疗前后△熵在预测有效中具有一定效能（AUC=0.673， P<0.05）。借助DCE-MRI纹理分析指标在判断乳腺癌患者NAC治疗后疗效中具有一定的应用价值。

abstractsThe purpose of this study was to investigate the value of MRI texture analysis in predicting the treatment response to neoadjuvant chemotherapy (NAC) in patients with breast cancer. The treatment response was assessed in 126 breast cancer patients at two weeks after NAC. The differences in MRI texture descriptors (inverse difference moment, entropy, energy and correlation) between baseline and after treatment were analyzed in patients according to treatment response to NAC. The value of these differences in predicting treatment response to NAC for breast cancer was determined. This study showed good performance with post-treatment entropy, energy and correlation (AUC=0.769, 0.753, 0.700, P<0.001) , and also with entropy difference between baseline and after treatment (AUC=0.673, P<0.05) , in predicting the treatment response to NAC. In summary, descriptors derived from DCE-MRI texture analysis may help evaluate the treatment response to NAC in patients with breast cancer.