1 ABSTRACT

Breast cancer is one of the most prevalent forms of cancer in the world. More than 250,000 American women are diagnosed with breast cancer annually. Fortunately, the survival rate is relatively high and continually increasing due to improved detection techniques and treatment methods. The quality of life of breast cancer survivors is ameliorated by minimizing adverse effects on their physical appearance. Breast reconstruction is important for restoring the survivor’s appearance. In breast reconstructive surgery, there is a need to develop technologies for quantifying surgical outcomes and understanding women’s perceptions of changes in their appearance. Methods for objectively measuring breast anatomy are needed in order to help breast cancer survivors, radiation oncologists, and surgeons quantify changes in appearance that occur with different breast reconstructive surgical options. In this study, we present an automated method for computing a variant of the normalized Breast Retraction Assessment (pBRA), a common measure of symmetry, from routine clinical photographs taken to document breast cancer treatment procedures.

Accurate assessment of the degree of scaring that results from surgical intervention for breast cancer would enable more effective pre-operative counseling. The resultant scar that accompanies an open surgical intervention may be characterized by variance in thickness, color, and contour. These factors significantly impact the overall appearance of the breast. A number of studies have addressed the mechanical and pathologic aspects of scarring. The majority of these investigations have focused on the physiologic process of scar formation and means to improve the qualities of a scar. Few studies have focused on quantifying the visual impact of scars. This manuscript critically reviews current methods used to assess scars in terms of overall satisfaction after surgery. We introduce objective, quantitative measures for assessing linear breast surgical scars using digital photography. These new measurements of breast surgical scars are based on calculations of contrast and area. We demonstrate, using the intra-class correlation coefficient (ICC), that the new measures are robust to observer variability in annotating the scar region on clinical photographs. As an example of the utility of the new measures, we use them to quantify the aesthetic differences of reconstruction following skin-sparing mastectomy vs. conventional mastectomy.

A good aesthetic outcome is an important endpoint of breast cancer treatment. Subjective ratings, direct physical measurements, measurements on photographs, and assessment by three-dimensional imaging are reviewed and future directions in aesthetic outcome measurements are discussed. Qualitative, subjective scales have frequently been used to assess aesthetic outcomes following breast cancer treatment. However, none of these scales has achieved widespread use because they are typically vague and have low intra- and inter- observer agreement. Anthropometry is not routinely performed because it is impractical to conduct the large studies needed to validate anthropometric measures, i.e., studies in which several observers measure the same subjects multiple times. Quantitative measures based on digital/digitized photographs have yielded acceptable results but have some limitations. Three-dimensional imaging has the potential to enable consistent, objective assessment of breast appearance, including properties, such as volume, that are not available from two-dimensional images. However, further work is needed to define 3D measures of aesthetic properties and how they should be interpreted.

Rationale, aims and objectives

Scarring is a significant cause of dissatisfaction for women who undergo breast surgery. Scar tissue may be clinically distinguished from normal skin by aberrant color, rough surface texture, increased thickness (hypertrophy), and firmness. Colorimeters or spectrophotometers can be used to quantitatively assess scar color, but they require direct patient interaction and can cost thousands of dollars By comparison, digital photography is already in widespread use to document clinical outcomes and requires less patient interaction. Thus, assessment of scar coloration by digital photography is an attractive alternative. The goal of this study was to compare color measurements obtained by digital photography and colorimetry.

Method

Agreement between photographic and colorimetric measurements of color were evaluated. Experimental conditions were controlled by performing measurements on artificial scars created by a makeup artist. The colorimetric measurements of the artificial scars were compared to those reported in the literature for real scars in order to confirm the validity of this approach. We assessed the agreement between the colorimetric and photographic measurements of color using a hypothesis test for equivalence, the intra-class correlation coefficient (ICC), and the Bland-Altman method.

Results

Overall, good agreement was obtained for three parameters (L*a*b*) measured by colorimetry and photography from the results of three statistical analyses.

Conclusion

Color measurements obtained by digital photography were equivalent to those obtained using colorimetry. Thus, digital photography is a reliable, cost-effective measurement method of skin color and should be further investigated for quantitative analysis of surgical outcomes.

Breast cancer is one of the most prevalent forms of cancer in the US. It is estimated that more than 180,000 American women will be diagnosed with invasive breast cancer in 2008. Fortunately, the survival rate is relatively high and continually increasing due to improved detection techniques and treatment methods. However, maintaining quality of life is a factor often under emphasized for breast cancer survivors. Breast cancer treatments are invasive and can lead to deformation of the breast. Breast reconstruction is important for restoring the survivor’s appearance. However, more work is needed to develop technologies for quantifying surgical outcomes and understanding women’s perceptions of changes in their appearance. A method for objectively measuring breast anatomy is needed in order to help both the breast cancer survivors and their surgeons take expected changes to the survivor’s appearance into account when considering various treatment options. In the future, augmented reality tools could help surgeons reconstruct a survivor’s breasts to match her preferences as much as possible.

The use of computer-aided detection (CAD) systems in mammography has been the subject of intense research for many years. These systems have been developed with the aim of helping radiologists to detect signs of breast cancer. However, the effectiveness of CAD systems in practice has sparked recent debate. In this commentary, we argue that computer-aided detection will become an increasingly important tool for radiologists in the early detection of breast cancer, but there are some important issues that need to be given greater focus in designing CAD systems if they are to reach their full potential.

Noise in mass spectrometry can interfere with identification of the biochemical substances in the sample. For example, the electric motors and circuits inside the mass spectrometer or in nearby equipment generate random noise that may distort the true shape of mass spectra. This paper presents a stochastic signal processing approach to analyzing noise from electrical noise sources (i.e., noise from instrumentation) in MALDI TOF mass spectrometry. Noise from instrumentation was hypothesized to be a mixture of thermal noise, 1/f noise, and electric or magnetic interference in the instrument. Parametric power spectral density estimation was conducted to derive the power distribution of noise from instrumentation with respect to frequencies. As expected, the experimental results show that noise from instrumentation contains 1/f noise and prominent periodic components in addition to thermal noise. These periodic components imply that the mass spectrometers used in this study may not be completely shielded from the internal or external electrical noise sources. However, according to a simulation study of human plasma mass spectra, noise from instrumentation does not seem to affect mass spectra significantly. In conclusion, analysis of noise from instrumentation using stochastic signal processing here provides an intuitive perspective on how to quantify noise in mass spectrometry through spectral modeling.

Purpose.

We identified candidate optical coherence tomography (OCT) markers for early glaucoma diagnosis. Time variation of retinal nerve fiber layer (RNFL) thickness, phase retardation, birefringence, and reflectance using polarization sensitive optical coherence tomography (PS-OCT) were measured in three non-human primates with induced glaucoma in one eye. We characterized time variation of RNFL thickness, phase retardation, birefringence, and reflectance with elevated intraocular pressure (IOP).

Methods.

One eye of each of three non-human primates was laser treated to increase IOP. Each primate was followed for a 30-week period. PS-OCT measurements were recorded at weekly intervals. Reflectance index (RI) is introduced to characterize RNFL reflectance. Associations between elevated IOP and RNFL thickness, phase retardation, birefringence, and reflectance were characterized in seven regions (entire retina, inner and outer rings, and nasal, temporal, superior and inferior quadrants) by linear and non-linear mixed-effects models.

Results.

Elevated IOP was achieved in three non-human primate eyes with an average increase of 13 mm Hg over the study period. Elevated IOP was associated with decreased RNFL thickness in the nasal region (P = 0.0002), decreased RNFL phase retardation in the superior (P = 0.046) and inferior (P = 0.021) regions, decreased RNFL birefringence in the nasal (P = 0.002) and inferior (P = 0.029) regions, and loss of RNFL reflectance in the outer rings (P = 0.018). When averaged over the entire retinal area, only RNFL reflectance showed a significant decrease (P = 0.028).

Conclusions.

Of the measured parameters, decreased RNFL reflectance was the most robust correlate with glaucomatous damage. Candidate cellular mechanisms are considered for decreased RNFL reflectance, including mitochondrial dysfunction and retinal ganglion cell apoptosis.

Measured time variation of retinal nerve fiber layer (RNFL) thickness, phase retardation, birefringence, and reflectance using polarization sensitive optical coherence tomography (PS-OCT) in three non-human primates with induced glaucoma in one eye over the course of 30 weeks.

Appearance changes resulting from breast cancer treatment impact the quality of life of breast cancer survivors, but current approaches to evaluating breast characteristics are very limited. It is challenging, even for experienced plastic surgeons, to describe how different aspects of breast morphology impact overall assessment of esthetics. Moreover, it is difficult to describe what they are looking for in a manner that facilitates quantification. The goal of this study is to assess the potential of using eye-tracking technology to understand how plastic surgeons assess breast morphology by recording their gaze path while they rate physical characteristics of the breasts, e.g., symmetry, based on clinical photographs. In this study, dwell time, transition frequency, dwell sequence conditional probabilities, and dwell sequence joint probabilities were analyzed across photographic poses and three observers. Dwell-time analysis showed that all three surgeons spent the majority of their time on the anterior–posterior (AP) views. Similarly, transition frequency analysis between regions showed that there were substantially more transitions between the breast regions in the AP view, relative to the number of transitions between other views. The results of both the conditional and joint probability analyses between the breast regions showed that the highest probabilities of transitions were observed between the breast regions in the AP view (APRB, APLB) followed by the oblique views and the lateral views to complete evaluation of breast surgical outcomes.

Principal-oscillation-pattern (POP) analysis is a multivariate and systematic technique for identifying the dynamic characteristics of a system from time-series data. In this study, we demonstrate the first application of POP analysis to genome-wide time-series gene-expression data. We use POP analysis to infer oscillation patterns in gene expression. Typically, a genomic system matrix cannot be directly estimated because the number of genes is usually much larger than the number of time points in a genomic study. Thus, we first identify the POPs of the eigen-genomic system that consists of the first few significant eigengenes obtained by singular value decomposition. By using the linear relationship between eigengenes and genes, we then infer the POPs of the genes. Both simulation data and real-world data are used in this study to demonstrate the applicability of POP analysis to genomic data. We show that POP analysis not only compares favorably with experiments and existing computational methods, but that it also provides complementary information relative to other approaches.

We evaluated the use of a stylus as a computer interface for radiographic image annotation. Our case study concerned the annotation of spiculated lesions on mammograms. Three experienced radiologists annotated 20 mammograms depicting spiculated lesions. We evaluated the interobserver agreement in annotations marked with a stylus versus those marked with a mouse using the intraclass correlation coefficient. Better agreement in annotating spicule width was observed with the stylus, suggesting that it is easier to accurately annotate subtle regions on an image using a stylus.

In this study we evaluate the influence of subject pose during image acquisition on quantitative analysis of breast morphology. Three (3D) and two-dimensional (2D) images of the torso of 12 female subjects in two different poses; (1) hands-on-hip (HH) and (2) hands-down (HD) were obtained. In order to quantify the effect of pose, we introduce a new measure; the 3D pBRA (Percentage Breast Retraction Assessment) index, and validate its use against the 2D pBRA index. Our data suggests that the 3D pBRA index is linearly correlated with the 2D counterpart for both of the poses, and is independent of the localization of fiducial points within a tolerance limit of 7 mm. The quantitative assessment of 3D asymmetry was found to be invariant of subject pose. This study further corroborates the advantages of 3D stereophotogrammetry over 2D photography. Problems with pose that are inherent in 2D photographs are avoided and fiducial point identification is made easier by being able to panoramically rotate the 3D surface enabling views from any desired angle.

The objective of this study was to determine if measurements of breast morphology computed from three-dimensional (3D) stereophotogrammetry are equivalent to traditional anthropometric measurements obtained directly on a subject using a tape measure. 3D torso images of 23 women ranged in age from 36 to 63 who underwent or were scheduled for breast reconstruction surgery were obtained using a 3dMD torso system (3Q Technologies Inc., Atlanta, GA). Two different types (contoured and line-of-sight distances) of a total of nine distances were computed from 3D images of each participant. Each participant was photographed twice, first without fiducial points marked (referred to as unmarked image) and second with fiducial points marked prior to imaging (referred to as marked image). Stereophotogrammetry was compared to traditional direct anthropometry, in which measurements were taken with a tape measure on participants. Three statistical analyses were used to evaluate the agreement between stereophotogrammetry and direct anthropometry. Seven out of nine distances showed excellent agreement between stereophotogrammetry and direct anthropometry (both marked and unmarked images). In addition, stereophotogrammetry from the unmarked image was equivalent to that of the marked image (both line-of-sight and contoured distances). A lower level of agreement was observed for some measures because of difficulty in localizing more vaguely defined fiducial points, such as lowest visible point of breast mound, and inability of the imaging system in capturing areas obscured by the breast, such as the inframammary fold. Stereophotogrammetry from 3D images obtained from the 3dMD torso system is effective for quantifying breast morphology. Tools for surgical planning and evaluation based on stereophotogrammetry have the potential to improve breast surgery outcomes.

Purpose

To determine the degree of correlation between spatial characteristics of the retinal nerve fiber layer (RNFL) birefringence (ΔnRNFL) surrounding the optic nerve head (ONH) with the corresponding anatomy of retinal ganglion cell (RGC) axons and their respective organelles.

Methods

RNFL phase retardation per unit depth (PR/UD, proportional to ΔnRNFL) was measured in two cynomolgus monkeys using enhanced polarization-sensitive optical coherence tomography (EPS-OCT). The monkeys were perfused with glutaraldehyde and eyes were enucleated and prepared for transmission electron microscopy (TEM) histological analysis. Morphological measurements from TEM images were used to estimate values of neurotubule density (ρRNFL), axoplasmic area (Ax) mode, axon area (Aa) mode, slope (u) of neurotubule number versus axoplasmic area [neurotubule packing density], fractional area of axoplasm in the nerve fiber bundle (f), mitochondrial fractional area in the nerve fiber bundle (xm), mitochondriated axon profile fraction (mp), and length of axonal membrane profiles per unit nerve fiber bundle area (Lam/Ab). Registered PR/UD and morphological parameters from corresponding angular sections were then correlated using Pearson’s correlation and multi-level models.

Results

In one eye, there was a statistically significant correlation between PR/UD and ρRNFL (r = 0.67, P =0.005) and between PR/UD and neurotubule packing density (r = 0.70, P = 0.002). Correlation coefficients of r = 0.81 (P=0.01) and r = 0.50 (P = 0.05) were observed between PR/UD and (Ax) mode for each respective subject.

Conclusion

Neurotubules are the primary source of birefringence in the RNFL of the primate retina.

Retinal nerve fiber layer (RNFL) thickness, a measure of glaucoma progression, can be measured in images acquired by spectral domain optical coherence tomography (OCT). The accuracy of RNFL thickness estimation, however, is affected by the quality of the OCT images. In this paper, a new parameter, signal deviation (SD), which is based on the standard deviation of the intensities in OCT images, is introduced for objective assessment of OCT image quality. Two other objective assessment parameters, signal to noise ratio (SNR) and signal strength (SS), are also calculated for each OCT image. The results of the objective assessment are compared with subjective assessment. In the subjective assessment, one OCT expert graded the image quality according to a three-level scale (good, fair, and poor). The OCT B-scan images of the retina from six subjects are evaluated by both objective and subjective assessment. From the comparison, we demonstrate that the objective assessment successfully differentiates between the acceptable quality images (good and fair images) and poor quality OCT images as graded by OCT experts. We evaluate the performance of the objective assessment under different quality assessment parameters and demonstrate that SD is the best at distinguishing between fair and good quality images. The accuracy of RNFL thickness estimation is improved significantly after poor quality OCT images are rejected by automated objective assessment using the SD, SNR, and SS.

Surgical reconstruction of natural-appearing breasts is a challenging task. Currently, surgical planning is limited to the surgeon’s subjective assessment of breast morphology. Therefore, it is useful to develop objective measurements of breast contour. In this paper, a novel quantitative measure of the breast contour based on catenary theory is introduced. A catenary curve is fitted on the breast contour (lateral and inferior) and the key parameter determining the shape of the curve is extracted. The new catenary analysis was applied to pre- and post-operative clinical photographs of women who underwent tissue expander/implant (TE/Implant) reconstruction. A logistic regression model was developed to predict the probability that the observed contour is that of a TE/Implant reconstruction from the catenary parameter, patient age, and patient body mass index. It was demonstrated that the parameters contain useful information for distinguishing TE/Implant reconstructed breasts from pre-operative breasts.

We have developed an algorithm for enhancement of spicules of spiculated masses, which uses the discrete radon transform. Previously, we employed a commonly used method to compute the discrete radon transform, which we refer to as the DRT. Recently, a new, more exact method to compute the discrete radon transform was developed by Averbuch et al, which is called the fast slant stack (FSS) method. Our hypothesis was that this new formulation would help to improve our enhancement algorithm. To test this idea, we conducted multiple two-alternative-forced-choice observer studies and found that most observers preferred the enhanced images generated with the FSS method.

Methods for identifying differential expression were compared on time series microarray data from artificial gene networks. Identifying differential expression was dependent on normalization and whether the background was removed. Loess after background correction improved results for most methods. On data without background correction median centering improved performance. We recommend Cui and Churchill’s ANOVA variants on background subtracted data and Efron and Tibshirani’s Empirical Bayes Wilcoxon Rank Sum test when the background cannot be removed.

Noise reduction is a critical step in proteomic profiling by mass spectrometry for identification of disease biomarkers. We propose a new method for characterizing chemical noise in MALDI TOF mass spectrometry using wavelet analysis of multiple noise realizations. The use of multiple measurements of the noise process enables a more reliable characterization than can be obtained from a single measurement. We analyzed the distributions and summary statistics of the wavelet coefficients.

We propose a new feature selection algorithm, Guilt-By-Association (GBA), which uses hierarchical clustering based on feature correlations to eliminate redundant features. GBA can be used in conjunction with other algorithms to produce a feature selection routine that explicitly considers both the similarities between features and their individual discriminatory powers. In this preliminary study, a simple form of GBA was investigated on simulated proteomic data.

Our long-term goal is to develop decision aids that will improve breast cancer treatment by explicitly taking aesthetics in the consideration. Essentially all breast cancer treatment involves surgery, which inevitably leaves scars. However, the extent and type of scarring is not the same for different surgeries (e.g., different forms of reconstruction.) We present our preliminary experiences in using image processing techniques to quantify scar characteristics in clinical photographs.

Quantitative, objective measurements of breast curvature computed from clinical photographs could be used to investigate factors that impact reconstruction and facilitate surgical planning. This paper introduces a novel quantitative measure of breast curvature based on catenary. A catenary curve is used to approximate the overall curvature of the breast contour, and the curvature measure is extracted from the catenary curve. The catenary curve was verified by comparing its length, the area enclosed by the curve, and the curvature measure from the catenary curve to those from manual tracings of the breast contour. The evaluation of the proposed analysis employed untreated and postoperative clinical photographs of women who were undergoing tissue expander/implant (TE/Implant) reconstruction. Logistic regression models were developed to distinguish between the curvature of breasts undergoing TE/Implant reconstruction and that of untreated breasts based on the curvature measure and patient variables (age and body mass index). The relationships between the curvature measures of untreated breasts and patient variables were also investigated. The catenary curve approximates breast curvature reliably. The curvature measure contains useful information for quantifying the curvature differences between breasts undergoing TE/Implant reconstruction and untreated breasts, and identifying the effect of patient variables on the breast shape.