Classification of small lesions in dynamic breast MRI: eliminating the need for precise lesion segmentation through spatio-temporal analysis of contrast enhancement

Authors:
Mahesh B. Nagarajan;Markus B. Huber;Thomas Schlossbauer;Gerda Leinsinger;Andrzej Krol;Axel Wismüller
Affiliations:
Department of Imaging Sciences and Biomedical Engineering, University of Rochester, Rochester, USA 14627;Department of Imaging Sciences and Biomedical Engineering, University of Rochester, Rochester, USA 14627;Department of Radiology, Ludwig Maximilians Universität, Munich, Germany 80336;Department of Radiology, Ludwig Maximilians Universität, Munich, Germany 80336;Department of Radiology, SUNY Upstate Medical University, Syracuse, USA 13210;Department of Imaging Sciences and Biomedical Engineering, University of Rochester, Rochester, USA 14627 and Department of Radiology, Ludwig Maximilians Universität, Munich, Germany 80336
Venue:
Machine Vision and Applications
Year:
2013

Citing 3
Cited 2

Support-Vector Networks

Machine Learning
Pattern Classification (2nd Edition)

Pattern Classification (2nd Edition)
LIBSVM: A library for support vector machines

ACM Transactions on Intelligent Systems and Technology (TIST)

Machine learning in medical imaging

Machine Vision and Applications
Classification of small lesions on dynamic breast MRI: Integrating dimension reduction and out-of-sample extension into CADx methodology

Artificial Intelligence in Medicine

Quantified Score

Hi-index	0.00

Visualization

Abstract

Characterizing the dignity of breast lesions as benign or malignant is specifically difficult for small lesions; they do not exhibit typical characteristics of malignancy and are harder to segment since margins are harder to visualize. Previous attempts at using dynamic or morphologic criteria to classify small lesions (mean lesion diameter of about 1 cm) have not yielded satisfactory results. The goal of this work was to improve the classification performance in such small diagnostically challenging lesions while concurrently eliminating the need for precise lesion segmentation. To this end, we introduce a method for topological characterization of lesion enhancement patterns over time. Three Minkowski Functionals were extracted from all five post-contrast images of 60 annotated lesions on dynamic breast MRI exams. For each Minkowski Functional, topological features extracted from each post-contrast image of the lesions were combined into a high-dimensional texture feature vector. These feature vectors were classified in a machine learning task with support vector regression. For comparison, conventional Haralick texture features derived from gray-level co-occurrence matrices (GLCM) were used. A new method for extracting thresholded GLCM features was also introduced and investigated here. The best classification performance was observed with Minkowski Functionals area and perimeter, thresholded GLCM features f8 and f9, and conventional GLCM features f4 and f6. However, both Minkowski Functionals and thresholded GLCM achieved such results without lesion segmentation while the performance of GLCM features significantly deteriorated when lesions were not segmented ( $$p ). This suggests that such advanced spatio-temporal characterization can improve the classification performance achieved in such small lesions, while simultaneously eliminating the need for precise segmentation.