Original contributionEvaluation of the three-time-point method for diagnosis of breast lesions in contrast-enhanced MR mammography
Introduction
Contrast-enhanced MR mammography is a useful complementary method to conventional diagnostic techniques on certain clinical indications. The clinical applicability of MR mammography is based on the advantages of its very high sensitivity on breast cancer (above 90%), but in contrast, the literature reports varying results concerning specificity (37–86%) [1], [2], [3], [4], [5], [6].
During the last years, a wide range of contrast-enhanced protocols and interpretation criteria were developed and evaluated with the attempt to increase the specificity and standardization of MR mammography [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15]. In addition, Degani et al. [9] and Furman-Haran et al. [10] developed a parametric method for diagnosis breast lesions. This method is based on using high-spatial-resolution images while scanning the images at judiciously selected three time points (one precontrast and two postcontrast time points). The enhancement rate defined by the intensity difference in the first two time points is coded by color intensity, and the change in enhancement between the second and third time points is coded by color hue (red, green, blue).
This method may provide an automated means of beast lesion identification and classification and should enable to more reliably differentiate benign vs. malignant breast lesions [12]. However, the 3TP method also enables an automated and pixel-by-pixel display of the lesion's contrast enhancement pattern. It was our hypothesis that the 3TP method may allow to avoid manual region-of-interest (ROI) placements that are time-consuming and have a large interreader variability and subjectivity.
The aim of our prospective study is to evaluate the three-time-point (3TP) method for its ability to produce equivalent results to manual ROI placement in terms of dynamic enhancement pattern for the diagnosis of breast lesions in contrast-enhanced MR mammography.
Section snippets
Patients
MR imaging was performed in 40 female patients (age range, 32–78 years; mean, 52 years) who were seen in our radiology department for mammographic and/or sonographic abnormalities.
MR imaging
MR imaging was performed on a 1.5-T scanner (Magnetom Sonata; Siemens, Erlangen, Germany), using a dedicated receive-only breast coil (phased-array breast coil; Siemens).
A three-dimensional axial gradient-echo acquisition was performed using the following parameters: TR/TE, 11/4,76 ms; flip angle, 15°; field of view,
Results
We included in the analysis 120 lesions in 40 patients. The average number of lesions was three with a range of 0–11. In 19 patients, lesions were also present in the contralateral breast.
Forty (33%) lesions had a diameter of less than 5 mm, 56 (47%) lesions 5–10 mm, and 24 (20%) lesions were greater than 1 cm (range, 0.3–3.7 cm; mean, 0.8 cm).
Of all 120 lesions, 65 (54%) showed heterogeneous color hue and color intensity with mixed blue, green, or red pixels. Twenty-one of these lesions were
Discussion
Time-intensity curves by manual placement of ROI are known to be useful in the measurement of dynamic enhancement of signal intensity over time following injection of contrast agent [5], [7]. The concentration change over time is predominantly determined by two pathophysiologic parameters that differ in malignant tumors and benign lesions. These parameters are the product to the blood vessel surface area (often referred to as vascular permeability) and the extracellular volume fraction
Conclusions
- 1.
The 3TP method automatically and reliably converts contrast kinetic information of the entire breast into a color-coded image.
- 2.
The 3TP method presents kinetic information of the entire dynamic series in an easy-to-interpret format and this automated method may allow to forego time-consuming and sometimes subjective manual ROI placements.
- 3.
With the 3TP method the entire lesion was analyzed on a pixel-by-pixel basis. Thus, this method displays the heterogeneity of the contrast enhancement pattern
References (15)
- et al.
MRI of the breast: histopathologic correlation
Eur J Radiol
(1987) - et al.
Assessment of breast cancer recurrence with contrast-enhanced subtraction MR imaging: preliminary results in 26 patients
Radiology
(1993) - et al.
Prognostic value of contrast-enhanced MR mammography in patients with breast cancer
Eur Radiol
(1997) - et al.
Application of a quantitative model to differentiate benign from malignant lesions detected by dynamic, gadolinium-enhanced MRI
J Magn Reson Imaging
(1996) - et al.
Dynamic breast MR imaging: are signal intensity time course data useful for differential diagnosis of enhancing lesions?
Radiology
(1999) MRI of breast tumors
Eur Radiol
(2000)- et al.
Signalverhalten maligner und benigner Läsionen in der dynamischen 2D-MRT der Mamma
Röfo
(1993)
Cited by (39)
An approach to the prediction of breast cancer response to neoadjuvant chemotherapy based on tumor habitats in DCE-MRI images
2023, Expert Systems with ApplicationsBreast DCE-MRI segmentation for lesion detection using Chimp Optimization Algorithm
2022, Expert Systems with ApplicationsCitation Excerpt :The analysis of clinically relevant regions of interest showed both a decrease of error (15%–62% reduction following registration) in tissue time-intensity curves and improved Areas Under the Curve (AUC60) at early enhancement. Hauth et al. (2006) described a method for voxel-wise curve categorization of the entire tumor, including voxel washing and washout. On a pixel-by-pixel basis, Three-Time Point (3TP) software translates the contrast kinetics into color-coded graphics.
Breast DCE-MRI segmentation for lesion detection by multi-level thresholding using student psychological based optimization
2021, Biomedical Signal Processing and ControlPathophysiological mapping of tumor habitats in the breast in DCE-MRI using molecular texture descriptor
2019, Computers in Biology and MedicineA geometrical perspective on the 3TP method in DCE-MRI
2015, Biomedical Signal Processing and ControlCitation Excerpt :The three time points and the pseudo-colour mapping were chosen in order both to emphasise the benign, malignant or uncertain nature of the tissue (which is derived from the time-course information) and to optimise the spatial resolution in breast DCE-MRI. Since then the 3TP method has been largely used in DCE-MRI clinical literature [2,6,8,18]. The fact that only three time-points are used to catch the information contained in the TCC makes the method sensitive to noise: one could argue that moderate noise levels superimposed to DCE images can cause benign/malignant tissues to be erroneously assigned to the uncertain class or vice-versa, while higher level of noise could lead to benign tissues confused with malignant and vice-versa.