Volumetric texture analysis of breast lesions on contrast‐enhanced magnetic resonance images

TLDR

Automated image analysis aims to extract relevant information from contrast‑enhanced magnetic resonance images of the breast and improve accuracy and consistency of interpretation. This study extends the 2D gray‑level co‑occurrence matrix method to a volumetric texture analysis approach for characterizing breast MR lesions. Using a 3D spoiled gradient‑echo T1‑weighted CE‑MRI database of 121 biopsy‑proven lesions, the authors segmented lesions with fuzzy c‑means clustering, computed nondirectional GLCMs from the first post‑contrast frame, extracted texture features, and evaluated their discriminative power with ROC analysis. Volumetric texture features achieved significantly higher classification performance than 2D analysis, and parameter optimization improved diagnostic accuracy. Published in Magn Reson Med 58:562–571 (2007) © 2007 Wiley‑Liss, Inc.

Abstract

Abstract Automated image analysis aims to extract relevant information from contrast‐enhanced magnetic resonance images (CE‐MRI) of the breast and improve the accuracy and consistency of image interpretation. In this work, we extend the traditional 2D gray‐level co‐occurrence matrix (GLCM) method to investigate a volumetric texture analysis approach and apply it for the characterization of breast MR lesions. Our database of breast MR images was obtained using a T1‐weighted 3D spoiled gradient echo sequence and consists of 121 biopsy‐proven lesions (77 malignant and 44 benign). A fuzzy c‐means clustering (FCM) based method is employed to automatically segment 3D breast lesions on CE‐MR images. For each 3D lesion, a nondirectional GLCM is then computed on the first postcontrast frame by summing 13 directional GLCMs. Texture features are extracted from the nondirectional GLCMs and the performance of each texture feature in the task of distinguishing between malignant and benign breast lesions is assessed by receiver operating characteristics (ROC) analysis. Our results show that the classification performance of volumetric texture features is significantly better than that based on 2D analysis. Our investigations of the effects of various of parameters on the diagnostic accuracy provided means for the optimal use of the approach. Magn Reson Med 58:562–571, 2007. © 2007 Wiley‐Liss, Inc.