Multispectral Co-Occurrence With Three Random Variables in Dynamic Contrast Enhanced Magnetic Resonance Imaging of Breast Cancer

Kale, Mehmet Cemil; Clymer, Bradley D.; Koch, R.; Heverhagen, Johannes T.; Sammet, Steffen; Stevens, Robert; Knopp, Michael V.

doi:10.1109/tmi.2008.922181

Cited by 10 publications

(11 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, a fourdimensional co-occurrence texture analysis approach (considering signal intensity variation over time) [22] and a multispectral co-occurrence analysis with three random variables (defined by three pharmacokinetic parameters) [23] were proposed for voxel classificationbased segmentation of the malignant breast tissue. These approaches yielded promising results.…”

mentioning

confidence: 99%

Assessing heterogeneity of lesion enhancement kinetics in dynamic contrast-enhanced MRI for breast cancer diagnosis

Karahaliou¹,

Vassiou²,

Arikidis³

et al. 2010

BJR

124

117

View full text Add to dashboard Cite

ABSTRACT. The current study investigates the feasibility of using texture analysis to quantify the heterogeneity of lesion enhancement kinetics in order to discriminate malignant from benign breast lesions. A total of 82 biopsy-proven breast lesions (51 malignant, 31 benign), originating from 74 women subjected to dynamic contrastenhanced magnetic resonance imaging (DCE-MRI) were analysed. Pixel-wise analysis of DCE-MRI lesion data was performed to generate initial enhancement, post-initial enhancement and signal enhancement ratio (SER) parametric maps; these maps were subsequently subjected to co-occurrence matrix texture analysis. The discriminating ability of texture features extracted from each parametric map was investigated using a least-squares minimum distance classifier and further compared with the discriminating ability of the same texture features extracted from the first post-contrast frame. Selected texture features extracted from the SER map achieved an area under receiver operating characteristic curve of 0.922 ¡ 0.029, a performance similar to post-initial enhancement map features (0.906 ¡ 0.032) and statistically significantly higher than for initial enhancement map (0.767 ¡ 0.053) and first post-contrast frame (0.756 ¡ 0.060) features. Quantifying the heterogeneity of parametric maps that reflect lesion washout properties could contribute to the computer-aided diagnosis of breast lesions in DCE-MRI. Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) significantly complements mammography and is characterised by its high sensitivity in detecting breast cancer. However, its specificity in distinguishing malignant from benign lesions is highly varied: reported values range from 37 up to 90% [1]. This variation is mainly due to the variety of image acquisition protocols and interpretation schemes adopted in clinical practice [2].Diagnostic criteria in DCE-MRI of breast masses [3, 4] include assessment of morphological features such as lesion shape, margin and enhancement homogeneity (internal architecture), as well as analysis/assessment of signal intensity-time curves generated from manually selected regions of interest (ROIs) within the lesion area. The analysis of signal intensity-time curves can be performed qualitatively (i.e. visual inspection of the curve shape [5,6]), by means of empirical parameters (e.g. relative enhancement, time-to-peak enhancement, washout ratio [7]) or quantitatively through pharmacokinetic modelling techniques [8].The subjective selection of ROI within the lesion accounts for the increased intra-and interobserver variability in the interpretation of lesion enhancement kinetics and for the discrepancy of reported findings [9,10]. While the selection of an ROI that captures the entire lesion is less subjective [11], it provides average enhancement kinetics estimates and completely ignores the heterogeneity of tumour vascular characteristics, which is diagnostically important [12]. Pixel-wise analysis of enhancement kinetics (based either on pharmacokinetic ...

show abstract

mentioning

confidence: 99%

Assessing heterogeneity of lesion enhancement kinetics in dynamic contrast-enhanced MRI for breast cancer diagnosis

Karahaliou¹,

Vassiou²,

Arikidis³

et al. 2010

BJR

124

117

View full text Add to dashboard Cite

show abstract

“…on individual subjects, probability thresholds cannot be applied universally. In contrast, the results of Kale et al (2008) indicated that a threshold of 40% -50% is appropriate in general.…”

Section: Resultsmentioning

confidence: 82%

“…Kale et al (2008) showed that a classifier trained and evaluated using axially-acquired data achieved a similar level of performance when evaluated on unseen sagitally-acquired data. This robustness likely stems from the fact that the algorithm assesses the co-occurence of fitted Brix model parameters for neighbouring voxels, rather than assessing spatial co-occurrence as in Fusco & Sansone (2012).…”

Section: Methods Based On Voxel-wise Classificationmentioning

confidence: 87%

“…The multispectral co-occurrence method of Kale et al (2008) is the only one of those listed in Table 3.1 that has been shown to possess any robustness to variations in MR protocol. Kale et al (2008) showed that a classifier trained and evaluated using axially-acquired data achieved a similar level of performance when evaluated on unseen sagitally-acquired data.…”

Section: Methods Based On Voxel-wise Classificationmentioning

confidence: 99%

“…This section presents an extension that addresses a weakness of the method of Kale et al (2008), in that the co-occurence array has many elements but only a few observations. In their 3.2 Method 1: 3D segmentation by distance-weighted 3 variable multi-spectral co-occurence 49 work, co-occurence matrices were computed based on a neighbourhood around voxels of size 5 ⇥ 5 ⇥ 2 (50 entries).…”

Section: Methods 1: 3d Segmentation By Distance-weighted 3 Variable Mumentioning

confidence: 99%

See 2 more Smart Citations

Computer assisted detection and characterisation of breast cancer in MRI

McClymont¹

View full text Add to dashboard Cite

This thesis presents a novel set of CAD (computer assisted detection/diagnosis) tools to assist the radiologist with the increasingly complex task of detecting and characterising breast lesions using magnetic resonance imaging (MRI). Commercial CAD systems presently fall short of automatically locating and classifying malignant lesions. Instead they automate many of the image processing and analysis functions that would otherwise have to be performed manually and visualise the data to aid interpretation. It is perhaps not surprising, therefore, that a recent meta-study concluded that existing breast MRI CAD does not improve the sensitivity and specificity of experienced radiologists and their interpretation remains essential. A recent review of breast MRI and MR spectroscopy concluded that what is needed are "quantitative features extracted preferably from the automatically segmented 3D lesion" and a more comprehensive assessment of lesions based on features/measurements "derived from MR multi-parametric acquisitions". This then was the motivation for the objectives of this thesis: (i) to develop an automatic 3D lesion segmentation algorithm for multi-modal breast MRI data; (ii) to develop features that quantitatively characterise the lesion (morphology, microvasculature, and microstructure) and other breast cancer signs from this data; and (iii) to evaluate these CAD tools using clinical breast MRI data.With regard to (i) a novel fully automatic method for segmentation (i.e., detection and delineation) of suspicious tissue in breast MRI is presented and evaluated. The method is based on mean-shift clustering and graph-cuts on a region adjacency graph. To the author's knowledge it is the first fully automatic method for breast lesion detection and delineation in breast MRI. The method was tested on a total of 102 lesions from two different vendors' scanner systems. The regions of interest identified by the method were compared with the ground truth (manually delineated by an experienced radiographer) and the detection and delineation accuracies quantitatively evaluated. One hundred percent of the lesions were detected with a mean of 4.5 ± 1.2 false positives per subject. This false-positive rate is nearly 50% better than previously reported for a fully automatic breast lesion detection system. The median Dice coefficient was ii 0.76 (interquartile range, 0.17), and 0.75 (interquartile range, 0.16) for the two scanner systems respectively.With regard to (ii) several new features for breast MRI CAD-derived from anatomical T1w and T2w images, DCE-MRI, and DW-MRI-are presented. They include features that characterise vascularity, blooming, and centripetal/centrifugal enhancement; and features extracted from the repartition of a lesion into mean-shift clusters. A new method for the fully automatic segmentation and measurement of the internal mammary vessels is also presented and evaluated. This was motivated by recent findings that the vascular cross-sectional area of internal mammary vessels is significantly...

show abstract

CADx Mammography

Costaridou¹

2010

Biomedical Image Processing

View full text Add to dashboard Cite

Multispectral Co-Occurrence With Three Random Variables in Dynamic Contrast Enhanced Magnetic Resonance Imaging of Breast Cancer

Cited by 10 publications

References 13 publications

Assessing heterogeneity of lesion enhancement kinetics in dynamic contrast-enhanced MRI for breast cancer diagnosis

Assessing heterogeneity of lesion enhancement kinetics in dynamic contrast-enhanced MRI for breast cancer diagnosis

Computer assisted detection and characterisation of breast cancer in MRI

CADx Mammography

Contact Info

Product

Resources

About