DCE MRI of breast lesions has high sensitivity (>90%) and reasonable specificity (>70%). However, the specificity for smaller lesions is known to be poorer due to the inability to confidently distinguish morphological and kinetic features. This work utilizes radiomics of model free parameter maps to improve the diagnostic accuracy of sub 1cm lesions with a PPV of 100% and a NPV of 91% obtained in a test dataset. The high level of correlation between texture features calculated from initial enhancement, overall enhancement, and area under the enhancement curve maps indicate that data acquisition beyond peak enhancement may be unnecessary.
This abstract and the presentation materials are available to members only; a login is required.