Daniel Chang

A new bias field correction method combining N3 and FCM for improved segmentation of breast density on MRI

PURPOSE Quantitative breast density is known as a strong risk factor associated with the development of breast cancer. Measurement of breast density based on three-dimensional breast MRI may provide very useful information. One important step for quantitative analysis of breast density on MRI is the correction of field inhomogeneity to allow an accurate segmentation of the fibroglandular tissue (dense tissue). A new bias field correction method by combining the nonparametric nonuniformity normalization (N3) algorithm and fuzzy-C-means (FCM)-based inhomogeneity correction algorithm is developed in this work. METHODS The analysis is performed on non-fat-sat T1-weighted images acquired using a 1.5 T MRI scanner. A total of 60 breasts from 30 healthy volunteers was analyzed. N3 is known as a robust correction method, but it cannot correct a strong bias field on a large area. FCM-based algorithm can correct the bias field on a large area, but it may change the tissue contrast and affect the segmentation quality. The proposed algorithm applies N3 first, followed by FCM, and then the generated bias field is smoothed using Gaussian kernal and B-spline surface fitting to minimize the problem of mistakenly changed tissue contrast. The segmentation results based on the N3+FCM corrected images were compared to the N3 and FCM alone corrected images and another method, coherent local intensity clustering (CLIC), corrected images. The segmentation quality based on different correction methods were evaluated by a radiologist and ranked. RESULTS The authors demonstrated that the iterative N3+FCM correction method brightens the signal intensity of fatty tissues and that separates the histogram peaks between the fibroglandular and fatty tissues to allow an accurate segmentation between them. In the first reading session, the radiologist found (N3+FCM > N3 > FCM) ranking in 17 breasts, (N3+FCM > N3 = FCM) ranking in 7 breasts, (N3+FCM = N3 > FCM) in 32 breasts, (N3+FCM = N3 = FCM) in 2 breasts, and (N3 > N3+FCM > FCM) in 2 breasts. The results of the second reading session were similar. The performance in each pairwise Wilcoxon signed-rank test is significant, showing N3+FCM superior to both N3 and FCM, and N3 superior to FCM. The performance of the new N3+FCM algorithm was comparable to that of CLIC, showing equivalent quality in 57/60 breasts. CONCLUSIONS Choosing an appropriate bias field correction method is a very important preprocessing step to allow an accurate segmentation of fibroglandular tissues based on breast MRI for quantitative measurement of breast density. The proposed algorithm combining N3+FCM and CLIC both yield satisfactory results.

Reduction of breast density following tamoxifen treatment evaluated by 3-D MRI: preliminary study.

This study analyzed the change in breast density in women receiving tamoxifen treatment using 3-D MRI. Sixteen women were studied. Each woman received breast MRI before and after tamoxifen. The breast and the fibroglandular tissue were segmented using a computer-assisted algorithm, based on T1-weighted images. The fibroglandular tissue volume (FV) and breast volume (BV) were measured and the ratio was calculated as the percent breast density (%BD). The changes in breast volume (ΔBV), fibroglandular tissue volume (ΔFV) and percent density (Δ%BD) between two MRI studies were analyzed and correlated with treatment duration and baseline breast density. The ΔFV showed a reduction in all 16 women. The Δ%BD showed a mean reduction of 5.8%. The reduction of FV was significantly correlated with baseline FV (P<.001) and treatment duration (P=.03). The percentage change in FV was correlated with duration (P=.049). The reduction in %BD was positively correlated with baseline %BD (P=.02). Women with higher baseline %BD showed more reduction of %BD. Three-dimensional MRI may be useful for the measurement of the small changes of ΔFV and Δ%BD after tamoxifen. These changes can potentially be used to correlate with the future reduction of cancer risk.

Computational simulation of breast compression based on segmented breast and fibroglandular tissues on magnetic resonance images

This study presents a finite element-based computational model to simulate the three-dimensional deformation of a breast and fibroglandular tissues under compression. The simulation was based on 3D MR images of the breast, and craniocaudal and mediolateral oblique compression, as used in mammography, was applied. The geometry of the whole breast and the segmented fibroglandular tissues within the breast were reconstructed using triangular meshes by using the Avizo 6.0 software package. Due to the large deformation in breast compression, a finite element model was used to simulate the nonlinear elastic tissue deformation under compression, using the MSC.Marc software package. The model was tested in four cases. The results showed a higher displacement along the compression direction compared to the other two directions. The compressed breast thickness in these four cases at a compression ratio of 60% was in the range of 5-7 cm, which is a typical range of thickness in mammography. The projection of the fibroglandular tissue mesh at a compression ratio of 60% was compared to the corresponding mammograms of two women, and they demonstrated spatially matched distributions. However, since the compression was based on magnetic resonance imaging (MRI), which has much coarser spatial resolution than the in-plane resolution of mammography, this method is unlikely to generate a synthetic mammogram close to the clinical quality. Whether this model may be used to understand the technical factors that may impact the variations in breast density needs further investigation. Since this method can be applied to simulate compression of the breast at different views and different compression levels, another possible application is to provide a tool for comparing breast images acquired using different imaging modalities--such as MRI, mammography, whole breast ultrasound and molecular imaging--that are performed using different body positions and under different compression conditions.

Impact of skin removal on quantitative measurement of breast density using MRI.

PURPOSE In breast MRI, skin and fibroglandular tissue commonly possess similar signal intensities, and as such, the inclusion of skin as dense tissue leads to an overestimation in the measured density. This study investigated the impact of skin to the quantitative measurement of breast density using MRI. METHODS The analysis was performed on the normal breasts of 50 women using nonfat-saturated (nonfat-sat) T1 weighted MR images. The skin was segmented by using a dynamic searching algorithm, which was based on the change in signal intensities from the background air (dark), to the skin (moderate), and then to the fatty tissue (bright). Tissue with moderate intensities that fell between the two boundaries determined based on the intensity gradients (from air to skin, and from skin to fat) was categorized as skin. The percent breast density measured with and without skin exclusion was compared. Also the relationship between the skin volume and the breast volume was investigated. Then, this relationship was used to estimate the skin volume from the breast volume for skin correction. RESULTS The percentage of the skin volume normalized to the breast volume ranged from 5.0% to 15.2% (median 8.6%, mean +/- STD 8.8 +/- 2.6%) among 50 women. The percent breast densities measured with skin (y) and without skin (x) were highly correlated, y = 1.23x+7% (r = 0.94, p < 0.001). The relationship between the skin volume and the breast volume was analyzed based on transformed data (the square root of the skin volume vs the cube root of breast volume) using the linear regression, and yielded r = 0.87, p < 0.001. When this model was used to estimate the skin volume for correction in the density analysis, it provided a better fit to the measured density with skin exclusion (with adjusted R2 = 0.98, and root mean square error = 1.6) compared to the correction done by using a fixed cutoff value of 8% (adjusted R2 = 0.83, root mean square error = 4.7). CONCLUSIONS The authors have shown that the skin volume is related to the breast volume, and this relationship may be used to correct for the skin effect in the MRI-based density measurement. A reliable quantitative density analysis method will aid in clinical investigation to evaluate the role of breast density for cancer risk assessment or for prediction of the efficacy of risk-modifying drugs using hormonal therapy.

Algorithm-based method for detection of blood vessels in breast MRI for development of computer-aided diagnosis.

To develop a computer‐based algorithm for detecting blood vessels that appear in breast dynamic contrast‐enhanced (DCE) magnetic resonance imaging (MRI), and to evaluate the improvement in reducing the number of vascular pixels that are labeled by computer‐aided diagnosis (CAD) systems as being suspicious of malignancy.

White matter lesion effect on tract-based spatial statistics for Alzheimer's patients

using positive and negative emotional words. Am.J.Psychiatry 160, 1938-1945 (2003). [5] Kelley, W. et al. Finding the self? An event-related fMRI study. J Cogn Neurosci 14, 785-794 (2002). [6] Moran, J., Heatherton, T. & Kelley, W. Modulation of cortical midline structures by implicit and explicit self-relevance evaluation. Soc.Neurosci 4, 197-211 (2009). [7] Giovacchini, G. et al. Brain incorporation of 11C-arachidonic acid, blood volume, and blood flow in healthy aging: a study with partial-volume correction. J. Nucl. Med 45, 1471-1479 (2004). [8] Ch etelat, G. et al. Direct voxel-based comparison between grey matter hypometabolism and atrophy in Alzheimer’s disease. Brain 131, 60-71 (2008).

Comparison of tract-based spatial statistics and ROI-based approach in analyzing the white matter integrity in the elderly

Background: As research continues to focus on the development of new treatments for Alzheimer’s disease (AD), and the selection of suitable subjects for clinical trials becomes increasingly important, the ability to reliably identify patients in the early or pre-symptomatic stages of the disease, and particularly those with amnestic mild cognitive impairment (aMCI), is desirable. The aim of this study is to classify subjects from the Alzheimer’s Disease Neuroimaging Initiative (ADNI), based on a regional analysis of their baseline and 12-month FDG-PET scans, as healthy controls (HC), or as having either aMCI or AD. Methods: Image data were obtained from 179 subjects (37 AD, 94 aMCI, 48 HC), whose baseline and 12-month FDG-PET scans were each re-aligned into the space of their corresponding baseline MRI, in which hippocampal masks were automatically generated for both timepoints. The signal intensity per cubic millimetre was determined in the hippocampus for both the baseline and 12-month FDG-PET scans, and the difference between the two calculated. Global variations in the cerebral metabolic rate of glucose between subjects were accounted for using the recently proposed ‘‘reference cluster’’ method, in which areas of apparent hypermetabolism between patients and controls (relatively unaffected by the disease) are extracted from the image data and used for normalisation. Results: Consideration of both baseline and follow-up data provides increased classification accuracies, determined using linear discriminant analysis, compared with those obtained using the baseline data alone. Accuracy increased from 72% to 78% between AD patients and HC, 65% to 68% between aMCI patients and HC, and 58% to 61% between AD and aMCI patients. Conclusions: This work-in-progress follows from the successful application of this regional analysis technique to baseline FDG-PET data from the ADNI, in which the features used for classification include not only the hippocampus, but all those extracted from a subject-specific segmentation consisting of 83 anatomical regions. These early results suggest that the group discrimination achieved using baseline data alone may be further improved by the inclusion of follow-up FDG-PET data.