Pseudocolour Image Processing in Digital Mammography

Transcription

1 Pseudocolour Image Processing in Digital Mammography M.Y. Sanavullah 1 and R. Samson Ravindran 2 1 Principal and Research Guide, VMKV Engineering College, V.M.R.F Deemed University, Salem, TN 2 Research scholar, VMKV Engineering College, V.M.R.F Deemed Univ ersity, Salem, TN Abstract Mammography is not a simple radiological technique alone, which is used to image breast tissues as it was known for the past two decades till Mammography is the leading method for breast imaging today. Pseudo coloring is applied for the 2D images taken in two plain film (X-ray) mammography at an angle 450 / 900 between them and the details are collected in digital form in a computer for segmentation and reconstruction. In this diagnostic process, Pseudo coloring plays a vital role in the detection of presumptive breast cancer or occult carcinoma. Keywords: Mammography, Breast imaging, 2D 3D digital imaging, Pseudo coloring 1. INTRODUCTION Digital mammography is a technique for recording x-ray images in computer code instead of on x-ray film, as with conventional mammography[3]. The images are displayed on a computer monitor and can be enhanced (lightened or darkened) before they are printed on film. Images can also be manipulated; the radiologist (a doctor who specializes in creating and interpreting pictures of areas inside the body) can magnify or zoom in on an area. The images can be stored and retrieved electronically, which makes long-distance consultations with other mammography specialists easier. Since the images can be adjusted by the radiologist, micro variations between tissues may be noted. The improved accuracy of digital mammography may reduce the number of follow up procedures. The first digital mammography system received U.S. Food and Drug Administration (FDA) approval in An example of a digital mammography system is the Senographe 2000D. 2. PSEUDO COLORING Pseudo color image processing shall be an important tool in digital image processing of breast images. Pseudo coloring comprises of assigning colors to gray values based on a specific criterion. The term Pseudo color or false color is used to differentiate the process of assigning colors to monochrome images from the process associated with true color images. The first and foremost use of Pseudo color is for human visualization and interpretation of gray scale events in an image or sequent of images. The principal cause of using color is the fact that humans can discern thousands if color shades and intensities, compared to only two dozen or shades of gray. 2.1 Density Slicing The technique of density also called intensity slicing and color coding is the simplest example of pseudo color image processing. If an image is interpreted as a 3D function[1][2], the method can be viewed as one of placing plan es parallel to the coordinate plane of the image. Each plane then slices the function in the area of intersection. Fig 1 shows a plane at f(x,y)=l i to slice the image function into two levels. I a different color is assigned to each side of the plane shown in Fig.1, any pixel whose gray scale is above the plane will be coded with one color, and any pixel below the plane will be coded with the other. The result is a two color image whose relative appearance can be controlled by moving the slicing plane up and down the gray level axis. The idea of planes is useful primarily for a geometric interpretation of the intensity-slicing technique. When more levels are used, the mapping function takes on a staircase form. 752

2 2.1.1 Gray Level to Color Transformations Fig. 1: Geometric interpretation of the intensity-slicing technique Fig 2: Functional block diagram for pseudo color image processing fr, fg, fb are fed into the corresponding Red, Green, and Blue inputs of an RGB color monitor There are many types of transformations and are capable of achieving a wider range of pseudo color enhancement results than the simple slicing technique discussed in the preceding section. An approach that is particularly attractive is shown in Fig.2. Basically the idea underlying this approach is to perform three independent transformations[6] on the gray level of any input pixel. The three results are then fed separately into the red, green, and blue channels of a color television monitor. This method produces a composite image, whose color content is modulated by the nature of the transformations on the gray-level values of an image, and is not functions of position Signal-to-noise Properties of Mammographic Film-screen Systems The Signal-to-Noise Ratio (SNR)[5][6] and the Detective Quantum Efficiency (DQE) have been experimentally determined as a function of spatial frequency for several mammographic film-screen systems. From the noise power spectra, it was found that film noise contributes significantly to the total noise of mammographic film-screen systems, comprising 30%-50% of the total noise at 1 cycle/mm and as much as 75% at 5 cycles/mm. All systems had approximately the same SNR below 1.5 cycles/mm, but differed at higher frequencies due to differences in screen MTF and in the gradient of the film's sensitometric curve. Generally, the DQE of mammographic film-screen systems is between 10%-30% at frequencies below 1 cycle/mm and decreases to about 1% between 8 and 12 cycles/mm. Compared to film-screen systems used in ge neral radiography, mammographic[9] systems have similar DQE values at low frequencies, but are superior at higher frequencies. 753

3 Pseudocolour Image Processing in Digital Mammography 3. IDENTIFICATION OF NORMAL DIGITAL MAMMOGRAMS Fig. 3: Steps for identifying the Normal Digital Mammograms 3.1 Process of Capturing Images for Digital Mammography Plain film mammography of a same patient are taken at 45 0 / 90 0 angle difference between the two images is shown in the Fig 3 and Fig 4. Fig. 3: (a) Cranio Caudal view of Right breast Fig. 3: (b) Medio Lateral view of right breast For detecting the micro calcifications in the above images, if the gray scale images are converted into color images then identifying calcium deposits or carcinoma will be easier. For this, the color policies available in the Photoshop software are utilized. 4. NEED FOR PSEUDO COLORING Pseudo coloring results in better identification of these calcifications[11]. The term Pseudo-color refers to using color to visualize things that aren t inherently colored. If two variables of information have to be displayed at a time in an image, the 754

4 intensity (one variable) and the hue (other variable) may be varied independently. Pseudo -color can be effectively used to label or identify particularly significant sections of the image as determined by some other image processing or vision algorithm. It has to be kept in mind while using a color that a large portion of the population (estimated 10% of all males) is color blind. If isoluminant displays are created (one with uniform intensity and varying hues), color blind people can t see the displays. Normal sighted people will see them poorly. 4.1 Color Slicing Highlighting a specific range of colors in an image is useful for separating objects from their surroundings. The basic idea is either to 1) display the colors of interest so that they stand out from the background or 2) use the region defined by the colors as a mask for further processing Image Compression Process Image compression[6][7] research aims at reducing the number of bits needed to represent an image by removing the spatial and spectral redundancies as much as possible. The foremost task is to find less correlated representation of the image. Two fundamental components of compression are redundancy and irrelevancy reduction. Redundancy reduction aims at removing duplication from the signal source (image/ video). Irrelevancy reduction omits parts of the signal that will not be noticed by the signal receiver, namely the Human visual system (HVS). Fig. 5: (a) Micro calcifications Fig. 5: (b) Microcalcifications after Pseudo coloring Fig. 6: (a) Circumscribed Lesions Fig. 6: (b) Obtained after applying color policies to circumscribed lesions Fig. 7: Obtained after applying pseudo color to circumscribed lesions 755

5 Pseudocolour Image Processing in Digital Mammography Fig. 8: Multi resolutions before and after Pseudo coloring Fig. 9: (a) Spiculated Lesions Fig. 9: (b) Spiculated lesions after Pseudo 5. RECONSTRUCTION Reconstruction[8] of position and the images are illustrated in Fig 10. The position can be detected by using two different x- ray tube positions (left image in Fig 10). The shape of the tissue can be reconstructed by using two spatially separated sources producing two different images depending on the density of the observed tissue (right image in the Fig 10) Fig. 10: Reconstruction of calcification position (left) and shape(right) using limited view reconstruction technique Position is reconstructed as intersection of projection lines for different x- ray tube angles. Shape is reconstructed by comparing the intensities in different projection images of the same calcification. Fig. 11: X ray position 756

6 1. For R CC CRANIO CAUDAL. 2. For R MLO MEDIO LATAERAL OBLIQUE The practical plain mammography R CC ( cranio caudal) is taken. In some different cases, we are analyzing the 45 0 right wards in the case of right breast[2] and 45 0 leftwards in the case of left breast as shown in the Fig 11. This is the basis for developing algorithm for wavelet transform Radon transform. The R MLO ( Medio Lataeral Oblique) is also taken. With these two available 45 0 separation details we have to simulate the 900 approximation. Using this we can get 2D image as shown in the Fig 7 which helps more to go for 3D[12] simulation using the transforms. (Wavelet transforms, direct cosine transforms, direct wavelet transforms and Radon transform). We can get a clear reconstruction of this digital image of the mammography in 3D. Fig.11: (a) Views Fig. 11: (b) Using 900 simulation arriving approximation manually Fi g. 12: Reconstruction of a single simulated calcification Fig. 13: Reconstructed single simulated calcification after Pseudo coloring An example of reconstructed single calcification is shown in the Fig.12. Using this approach, 3-D distributions of homogenous or linear malignant clusters and of peripheral plate like benign clusters are more clearly differentiated. This approach offers parametric representation of reconstructed calcifications, which can potentially reduce the amount of data needed for visualization, transmission and storing and can be of significance in telemammography [3] [4]. 6. CONCLUSION Thus the application of Pseudocolouring with the help of Photoshop colour policies is used to separate the carcinoma and to apply various colours for the calcifications in gray scale for better perception and understanding. The diagnostic quality of the X-ray based images using 3-D mammographic technique is better than those obtained by other modalities. We have found that the 3-D mammographic breast imagi ng techniques with the help of Pseudocoloring have potentials for both early cancer detection and diagnosis. REFERENCES [1] Maidment ADA, Albert. M, Conant. EF, Three Dimensional Imaging of Breast Calcifications, Proceeding on.spie, pp

7 Pseudocolour Image Processing in Digital Mammography [2] Kopans.DB(1998). Breast Imaging, 2nd Edition. Philadelphia :Lippincott- Raven. [3] Relijin B.Reljin I. In: Spasic P, Milosavlijevic I, Jancic Zguncas M,eds(2000). Text book Telemedicine, Belgrade Academy of Medical Sciences of Serbian Medical Association. [4] Nichole.F.R.Bolender s(2002), Hand book of science and Technology Mammography McGraw Hill Publications. [5] Rafael C.Gonzalez, Richard E.Woods Digital Image Processing Text book. [6] Pratt.W.K(2003). Digital image processing, John Wiley & sons. [7] S. Liu, C. F. Babbs, and E. J. Delp(1998), "Normal Mammogram Analysis and Recognition" Proceedings of the IEEE International Conference on Image Processing, pp.4-7. [8] S.Liu, C.F. Babbs, and E.J.Delp Multiresolution Detection of Spiculated Lesions in Digital Mammograms", IEEE Transactions on Image Processing. [9] Professor Edward J.Delp - Digital Mammography Research at Purdue University [10] B.Zheng, Y.H.Chang, and D.Gur(1995), Computerized Detection of masses in Digitized Mammograms using single image segmentation and a multilayer topograophic feature analysis, Acad. Radiology., Volume. 2. pp [11] Yam, M., Brady M., Highnam, R., Behrenbruch, C., (2001), Three dimensional Reconstruction of micro-calcifications clusters from two mammographic views, IEEE Transaction on Medical Imaging, Volume. 20(6), pp