Proc. SPIE. 10140, Medical Imaging 2017: Digital Pathology
KEYWORDS: Principal component analysis, Independent component analysis, Tissues, Image processing, Pathology, Multiplexing, Image analysis, Medical imaging, Image quality, Signal processing, Deconvolution, Absorbance, RGB color model
Computer based automatic medical image processing and quantification are becoming popular in digital pathology. However, preparation of histology slides can vary widely due to differences in staining equipment, procedures and reagents, which can reduce the accuracy of algorithms that analyze their color and texture information. To re- duce the unwanted color variations, various supervised and unsupervised color normalization methods have been proposed. Compared with supervised color normalization methods, unsupervised color normalization methods have advantages of time and cost efficient and universal applicability. Most of the unsupervised color normaliza- tion methods for histology are based on stain separation. Based on the fact that stain concentration cannot be negative and different parts of the tissue absorb different stains, nonnegative matrix factorization (NMF), and particular its sparse version (SNMF), are good candidates for stain separation. However, most of the existing unsupervised color normalization method like PCA, ICA, NMF and SNMF fail to consider important information about sparse manifolds that its pixels occupy, which could potentially result in loss of texture information during color normalization. Manifold learning methods like Graph Laplacian have proven to be very effective in interpreting high-dimensional data. In this paper, we propose a novel unsupervised stain separation method called graph regularized sparse nonnegative matrix factorization (GSNMF). By considering the sparse prior of stain concentration together with manifold information from high-dimensional image data, our method shows better performance in stain color deconvolution than existing unsupervised color deconvolution methods, especially in keeping connected texture information. To utilized the texture information, we construct a nearest neighbor graph between pixels within a spatial area of an image based on their distances using heat kernal in <i>l</i>αβ space. The representation of a pixel in the stain density space is constrained to follow the feature distance of the pixel to pixels in the neighborhood graph. Utilizing color matrix transfer method with the stain concentrations found using our GSNMF method, the color normalization performance was also better than existing methods.
It is a widely held belief that in the long run, three-dimensional (3D) display should supply stereo to multiple
viewers without wearing any viewing aids and free to move. Over the last few decades, great e®orts have
been made to approach auto-stereoscopic (AS) display for multiple viewers. Spatial multiplexing technique has
¯rst been employed to accommodate multiple viewers simultaneously in stereoscopic planar display. However,
resolution of each view image decreases as the number of viewers increases. Recent development of high-speed
liquid crystal display (LCD), which is capable of operating 240-Hz frame rate, makes feasible multi-viewer
display via time multiplexing and improving image quality at the same time. In this paper, we propose a
display adjustment algorithm that enables high-quality auto-stereoscopic display for multiple viewers. The
proposed method relies on spatio-temporal parallax barrier to channel desired stereo pair to corresponding
viewers according to their locations. We subsequently conduct simulations that demonstrate the e®ectiveness of
the proposed method.