Computerized analysis of breast lesions in three dimensions using dynamic magnetic‐resonance imaging

Abstract

Contrast-enhanced magnetic resonance imaging (MRI) of the breast is known to reveal breast cancer with higher sensitivity than mammography alone. The specificity is, however, compromised by the observation that several benign masses take up contrast agent in addition to malignant lesions. The aim of this study is to increase the objectivity of breast cancer diagnosis in contrast-enhanced MRI by developing automated methods for computer-aided diagnosis. Our database consists of 27 MR studies from 27 patients. In each study, at least four MR series of both breasts are obtained using FLASH three-dimensional (3D) acquisition at 90 s time intervals after injection of Gadopentetate dimeglumine (Gd-DTPA) contrast agent. Each series consists of 64 coronal slices with a typical thickness of 2 mm, and a pixel size of 1.25 mm. The study contains 13 benign and 15 malignant lesions from which features are automatically extracted in 3D. These features include margin descriptors and radial gradient analysis as a function of time and space. Stepwise multiple regression is employed to obtain an effective subset of combined features. A final estimate of likelihood of malignancy is determined by linear discriminant analysis, and the performance of classification by round-robin testing and receiver operating characteristics (ROC) analysis. To assess the efficacy of 3D analysis, the study is repeated in two-dimensions (2D) using a representative slice through the middle of the lesion. In 2D and in 3D, radial gradient analysis and analysis of margin sharpness were found to be an effective combination to distinguish between benign and malignant masses (resulting area under the ROC curve: 0.96). Feature analysis in 3D was found to result in higher performance of lesion characterization than 2D feature analysis for the majority of single and combined features. In conclusion, automated feature extraction and classification has the potential to complement the interpretation of radiologists in an objective, consistent, and accurate way.

Computer-Aided Diagnosis Scheme for Distinguishing Between Benign and Malignant Masses in Breast DCE-MRI

“…The objective features used for the QDA were selected from the 11 objective features according to a stepwise method Fig. 1 Example of dynamic contrast material-enhanced magnetic resonance imaging (DCE-MRI) images with a malignant mass after a duration of 0, 1, 2, and 5 min based on Wilks's lambda [22,23]. The stepwise method first selects the objective feature with the lowest Wilks's lambda for entry into the discriminant function.…”

Computer-Aided Diagnosis in Breast MRI: Do Adjunct Features Derived from T 2-weighted Images Improve Classification of Breast Masses?

“…The DCE T 1 -weighted features are circularity and variance of the radial gradient histogram as described in [5], reflecting the lesion shape and the sharpness of its margin, respectively. The three T 2 -weighted features are the normalized α-percentile for the eroded lesion segmentation (α = 20%) and the normalized α-percentile for the lesion shell with (α = 92%, t = 2mm) and (α = 98%, t = 20mm).…”

“…The sharpness of the lesion margin in the T 2 -weighted image is quantified by evaluating the mean and variance of the margin gradient as described for DCE T 1 -weighted images in [5] in the static T 2 -weighted images. These features reflect the mean and variance of gradient magnitudes computed with Gaussian derivatives of reasonable scale for all voxels in a three voxels thick shell centered on the lesion segmentation's surface.…”

“…A total of 22 morphological and kinetic T 1 -weighted features described in [5,6,7] were evaluated for all binary lesion segmentations. Due to the limited space we have to refer the reader to the corresponding articles for more information about these features.…”

“…(Semi-)automatic discrimination of benign and malignant masses in breast MRI was investigated by several authors [5,6,7]. The proposed approaches have in common that they solely concentrate on processing of kinetic and morphologic features as depicted in the DCE T 1 -weighted images, which are, without doubt, the most important source of information in breast MRI.…”