DCE-MRI data from 55 breast cancer patients collected before and after the first cycle of neoadjuvant chemotherapy were subjected to pharmacokinetic analysis. Four texture features, GLCM, RLM, single- and multi-resolution fractals extracted from DCE-MRI parametric maps, were analyzed for early prediction of therapy response. Generally, the multi-resolution fractal features from individual maps or the concatenated features from all parametric maps showed better predictive performance. The results suggest that multi-resolution analysis, which decomposes the texture at various spatial-frequency scales, may more accurately capture changes in tumor vascular heterogeneity as measured by DCE-MRI, and thus provide better early prediction of therapy response.
This abstract and the presentation materials are available to members only; a login is required.