Improving the discrimination of benign and malignant breast MRI lesions using the apparent diffusion coefficient
Paper i proceeding, 2010

This paper presents an investigation of the apparent diffusion coefficient (ADC) for improving the discrimination of benign and malignant lesions in breast magnetic resonance imaging (MRI). In particular a method is presented for automatically selecting hyper intense tumour voxels in dynamic contrast enhanced (DCE) MRI data and evaluating their average ADC in the corresponding diffusion-weighted (DW) MRI data. The method was applied to ten breast MRI datasets obtained from routine clinical practice. The results demonstrate that the combination of the relative signal increase (DCE-MRI) with the apparent diffusion coefficient (DW-MRI) leads to better discrimination than with either feature alone. The results also suggest that it is important to acquire the DW-MRI data in a consistent fashion, i.e. either before or after the acquisition of the DCE-MRI data.

breast cancer

contrast enhancement


apparent diffusion coefficient


Darryl McClymont

Andrew Mehnert

Chalmers, Signaler och system

Adnan Trakic

Stuart Crozier

Dominic Kennedy

Proc. 2010 International Conference on Digital Image Computing: Techniques and Applications (DICTA)

569 - 574


Livsvetenskaper och teknik


Datorseende och robotik (autonoma system)

Medicinsk bildbehandling