Classification of hypervascularized lesions in CE MR imaging of the breast

Abstract

The purpose of this paper is to define and evaluate a classification category for contrast-enhanced (CE) MR imaging of the breast based on the BI-RADS mammographic categories of the American College of Radiology. Using five evaluation criteria for MR findings (initial signal increase, post-initial signal behavior, shape, border, and contrast material distribution within enhancing tumors) 522 patients (1031 breasts) were analyzed. Scores were given from 0 to 8 points and classified into five categories (group I: 0 points, negative; group II: 1-2 points, benign; group III: 3 points, probably benign; group IV: 4-5 points, suspicious abnormality; group V: 6-8 points, highly suspicious for malignancy) to 265 focal hypervascularized breast lesions in 244 breasts (patient group A). These findings were correlated with histology or follow-up. Additionally, this classification was correlated to the contrast medium uptake within the parenchyma of the remaining 787 breasts without any focal lesion (patient group B). Two hundred sixty-five hypervascularized lesions in 238 patients (244 breasts, patient group A) were classified into group I: 0%; group II: 27.3%; group III: 22.3%; group IV: 18.6%; and group V: 31.8%. Histology revealed 115 benign and 134 malignant tumors in these groups. Sixteen benign lesions were controlled by follow-up. Sensitivity for the detection of malignancy using the presented multifactorial MRM classification was 92%, and specificity was 92%. Excluding cases of ductal carcinoma in situ specificity increased to 95%. Seven hundred eighty-seven breasts without any focal hypervascularized lesion (patient group B) were classified into groups I or II. Follow-up ( n=771) or histology ( n=14) confirmed the diagnosis in 785 of these breasts. Histopathology revealed, however, malignant tumors in the remaining two cases. The classification of lesions based on a multifactorial analysis is very helpful in the interpretation of CE MRI of the breast. The evaluation of all diagnostic imaging modalities, however, is essential in determining the correct diagnosis and/or in deciding on the appropriate therapeutic procedure.

Detection and classification of contrast‐enhancing masses by a fully automatic computer‐assisted diagnosis system for breast MRI

“…Invasive breast carcinomas may present central necrosis leading to encapsulated holes in the segmented ROIs; the presence of such holes was detected by a connectivity algorithm (25). While benign lesions often are sharply demarcated, malignant tumors frequently show a gradual transition in the surrounding tissue (9,11,32). The adjacent voxels in the neighborhood of the lesions were therefore also considered, and their average signal intensity was compared with the regions' intensities to assess the characteristic of the margins.…”

“…The initial phase was defined as the relative signal increase during the first two postcontrast sequences after contrast medium administration. Delayed or postinitial enhancement characterized the relative signal intensity change afterwards; three types of postinitial curves were defined: 1) continuous increase (signal increase >10%); 2) plateau phenomenon (stable signal intensity 610%); 3) signal decrease >10% as washout sign (6,9,19).…”

“…Numbers in parentheses are percentages. distinguishing between benign and malignant breast lesions (6,(9)(10)(11). Each of the descriptors alone is a specific, but not a very sensitive indicator of malignancy (6,9,10,32); for instance, 87% of enhancing lesions with washout sign have been shown to be malignant, but washout can be detected in only 57% of malignancies (6).…”

“…To improve specificity without significantly reducing sensitivity, morphologic features have been implemented in clinical routine as further diagnostic criteria in magnetic resonance mammography (MRM) (9)(10)(11). However, for interpreting dynamic as well as morphologic parameters, substantial interobserver variability has been reported (11)(12)(13).…”

Evaluation of Novel Genetic Algorithm Generated Schemes for Positron Emission Tomography (PET)/Magnetic Resonance Imaging (MRI) Image Fusion

“…In this context, F-18-FDG PET [2][3][4][5] and highresolution and dynamic contrast-enhanced MRI [6][7][8][9][10][11] have steadily gained acceptance in addition to X-ray mammography and ultrasonography. Initial experience with combined PET (functional imaging) and MRI (anatomical localization) has yielded positive results, providing benefits compared to either modality alone [12][13][14][15].…”