IMPROVEMENTS IN X-RAY CT

Transcription

1 The First International Proficiency Testing Conference Sinaia, România 11 th 13 th October, 2007 IMPROVEMENTS IN X-RAY CT Emilia Dana Seleţchi Faculty of Physics, University of Bucharest, Măgurele, CP MG-11 RO , Bucharest, Romania seletchi@gmail.com Abstract Computed Tomography (CT) is a diagnostic and research imaging procedure that uses X-rays, an elaborate radiation detection system and a computer to produce cross-sectional images of the patient s body. This paper deals with different methods for scanning, errors in the measurement of projection data and clinical applications of X-ray CT. Image enhancement techniques with Adobe Photoshop, ImageJ, Origin, etc., software have been used in order to increase the signal-to-noise ratio and accentuate the image features. Statistical functions help us to analyze the general characteristics of a CT image by displaying the image histogram and plotting the 3D Color Map Surface. Key words X-ray CT, Histogram, Skewness, Kurtosis, Plot Profile, 3D Color Map Surface 1 INTRODUCTION Computed Tomography (CT), originally known as Computed Axial Tomography (CAT) is a powerful nondestructive evaluation technique for producing 2-D or 3-D cross-sectional image of body tissues and organs. X-ray scans furnish detailed images of an object such as dimensions, shape, internal defects and density for diagnostic and research purposes. Supposing that we have a very narrow pencil beam of monochromatic X rays which traverse an inhomogeneous medium and no scattered radiation reaching the detector, the transmitted intensity is given by: I = I [ ] 0 exp μ x,y dl (1) l where I 0 is the unattenuated intensity, μ[x,y] is the two-dimensional distribution of the linear attenuation coefficient and l is the straight line joining the source and 401

2 detector [2, 3]. In CT scanners the X-ray attenuation according to equation (1) is measured along a variety of lines within a plane perpendicular to the long axis of the patient with the goal of reconstructing a map of the attenuation coefficients H for this plane [5]. The resulting attenuation coefficients, in Hounsfield units are usually expressed with reference to water: μtissue μwater H = 1000 (2) μwater Small differences in H can be amplified visually by increasing the contrast of the display.in a third generation fan beam x-ray tomography machine a point source of x-rays and a detector array are rotated continuously around the patient. Data collection time for such scanners ranges from 1 to 20 seconds. A special computer program calculates the values of density and creates cross-sectional images of the brain. Modern CT scanner can acquire data in a continuous helical or spiral fashion, shortening acquisition time and reducing artifacts. Beam hardening artifacts caused by polychromatic X-ray sources are most noticeable in the CT images. As the x-ray passes through tissue, the lower-energy components are attenuated more rapidly than the higher-energy components. Any attempt to reconstruct the image from the unprocessed data without making appropriate corrections for the beam hardening will result in error. These errors give rise to beamhardening artifacts, defects such as dark bands that are observable in the image. The non-linear relationship between the projection of the object, λ and μ causes the problem [1,6]: E max ( ) ( [ ] ) S E exp μ x,y dl de 0 l λ = ln E ( ) (3) max S E de 0 where the source spectrum S(E) is defined such that S(E) de is the energy fluence in the energy range E to E+dE. The main effects of this artifact show up as a false reduction in density in the centre of a uniform object and the creation of false detail in the neighborhood of bone/soft tissue interfaces. This problem is caused by the different paths lengths of tissue through which the projection are measured. Another problem is caused by the inhomogeneous nature of the object studied. The energy dependence of the linear attenuation coefficient varies according to the type of material: bone, water-equivalent tissue, air, etc. resulting additional artifacts into the image. A reduction of the noise level requires the increasing of x-ray intensity or data acquisition time, leading to the increase in patient dose. The radiation dose depends on multiple factors: volume scanned, patient build, number and type of scan sequences, desired resolution and image quality. X-ray CT, do expose the patient to a certain amount of radiation which should be the lowest in accordance with the radiation protection principle ALARA (as low as reasonable achievable) [4]. The typical effective dose for an X-ray CT brain scan (without contrast) is about 2-3 msv. X-ray scatter also leads to another type of error in the measurement of a projection. The most effective way to diminish nonlinear partial volume effects is to reduce the slice thickness. Image imperfections can also be caused by insufficient calibration of detector sensitivity, inadequacies in the reconstruction algorithm, non-uniformity scanning motion, fluctuation in x-ray tube voltage, etc. This paper presents the results concerning the X-ray CT image processing and analysis techniques by displaying histograms, Profile Plots, Power Spectra using Fast 402

3 Fourier Transformations (FFT) algorithm, 3D Color Surface Graphs and features of abnormal tissue growths. 2 EXPERIMENTAL 2.1 CT Instrumentation Computed Tomography uses an X-ray tube, an elaborate radiation detection system and a computer that assembles the measurement data into a series of transversal slice of the subject s body. There are two scan configurations that lead to rapid data collection. In a third-generation fan beam X-ray tomography machine, a multicellular detector system is rotated continuously around the patient together with the X-ray tube (Figure 1). The gantry is a frame housing the X-ray tube, collimators and detectors in a CT machine, with a large opening into the subject is inserted. The fourth-generation scanners use a stationary ring of detector and the fan shaped X-ray beam rotates around the patient. In spiral CT, the images are achieved by moving the subject slowly through the gantry while continuously acquiring data. Two different types of detector systems are mainly used by the CT scans. Sodium iodide detectors containing a single crystal of NaI with trace amounts of Tl (scintillation crystal) and rare-gas ionization chamber detectors. In fan beam systems a collimator is usually employed in front of the scintillator elements in order to diminish the effects of scattered X-rays. The detectors transmit the X-ray energy to a computer system which converts the radiation into electrical signals. Figure 1 - The CT equipment 1 - CT scanner, 2 - X-ray tube, 3- X-ray detector, 4,5 collimator, 6 - the couch (it can slide backwards and forwards through the hole of the CT scanner), 7- data acquisition unit, 8- computer monitor, 9- X-ray beam, 10- subject 2.2 Computer Software By using ImageJ, Adobe Photoshop 7.0, Corel PHOTO-PAINT 12.0 and OriginPro 7.5 software I have been realized the image processing and data analysis on X-ray CT images of normal and abnormal brain. ImageJ is a public domain Java image processing program. I have been used this software in order to measure distances, to calculate area and pixel value statistics of user-defined selections and to provide density histograms and line profile plots. Adobe Photoshop 7.0 image processing software has been used in conjunction with Corel PHOTO-PAINT 12.0 programs to improve the CT images by adjusting and creating special effects. OriginPro 7.5 is a 403

4 specialized program for data analysis providing FFT analysis, Profile Plots and 3D Color Maps Surface of CT images [7]. 3 RESULTS AND DISCUSSIONS 3.1 Image processing Image processing techniques can help to differentiate the abnormal tissue growths (tumors) in question from other tissues, providing more detailed information on head injuries, stroke, brain disease and internal structures than do regular X-ray CT scans [8]. By using suitable programs into the first stage I performed multiple processing on X-ray CT scans of an abnormal brain S1 (Subject 1) and S2 (Subject 2) illustrated in figure 2.a,b. Figure 2 - RGB-X-ray CT scan of (a) S1-abnormal brain ( pixels) where the subject s nose and ear lobes are clearly visible, (b) S2-abnormal brain where the subject s eyes are clearly visible ( pixels) The X-ray CT images of the brain were performed by using a Siemens Sensation 4 VA47 C. The Contrast Enhancement filter has been used to adjust the tone, color and contrast in the X-ray CT images. Figure 3 (a) 8-bit image performed by ImageJ triple processing: Enhance Contrast (Saturated Pixels 0.5 %, Equalize Histogram), Binary (Threshold) followed by Variance filter (Radius 6 pixels) on X-ray CT scan of S1 abnormal brain and (b) 8-bit image performed by Corel PHOTO-PAINT processing Transform, Threshold (Bi-level 180) of S2 abnormal brain. The Threshold setting changes pixel contrast, which can reduce or eliminate visible dust particles and other tiny marks. The radius setting enables you to control the number of pixels involved in the smoothing effect that is applied. Threshold 404

5 adjustment converts all colors to either black or white based on their brightness values (Fig. 3.a,b.). The Histogram Equalization filter was applied to redistribute the balance of shadows, midtones and highlights in the composite channel or in individual color channels. In order to highlight the edges in the X-ray CT images of normal brain I have been applied the Variance filter (radius 6) from ImageJ process menu. For clarity some regions are made transparent while the significant details can be easily seen. Adobe Photoshop filters used in conjunction with Corel PHOTO-PAINT processing enable to apply automated effects to an image, allowing us to correct lighting and perspective fluctuations, increasing the focus of an image and adding depth to RGB X-ray CT image. Psychedelic effect was used to shift an entire RGB image from one color range to another. Contour filters detect and accentuate the edges of objects and selections in the X-ray CT images of the brain. By using the Invert filter after Trace Contour process, every color in the X-ray CT image was converted to its exact opposite (Fig.4.a,b.). Hue represents color, saturation indicates the color depth or richness and lightness shows the overall percentage of white in the X-ray CT images. Figure 4 - Corel PHOTO-PAINT effects: Color Transform (Psychedelic) followed by Adobe Photoshop multiple filtering: Stylize (Trace Contour, Lower Edge followed by Find Edges) and image adjustments: Invert followed by Hue -Saturation -Lightness on X-ray CT scan of (a) S1 abnormal brain (b) S2 abnormal brain 3.2 Data analyze and interpretations ImageJ software was used to perform Histograms and Profile Plots for X-ray CT scans of S1 and S2 abnormal brain. Histogram illustrates the number of pixels distributed on X-ray CT image (y-axis) for each level (gray value) from darkest (0) to brightest (256). The total pixel count was also calculated and displayed, as well as the mean, modal, minimum and maximum gray value by using ImageJ program (Fig. 5.a,b). Figure 5 ImageJ histograms of (a) S1- X-ray CT abnormal brain scan (b) S2- X-ray CT abnormal brain scan 405

6 Count indicates the total number of pixels corresponding to the intensity level. Mean ( for S1 and for S2) shows the average intensity value. It is the sum of the gray values of all the pixels in the selection divided by the number of pixels. Std Dev (Standard Deviation) with the values and for S1 and S2 respectively, indicates how widely intensity values vary. Min (0) and Max (255 & 240) represents the minimum and maximum gray values within the X-ray CT images. Figure 5.b shows that there are no pure whites or pure blacks in the S2 X-ray CT abnormal brain scan. The Mode (Modal Gray Values: 5 & 18) was computed as the midpoint of the histogram interval with the highest peak. Particles Analyze command counts and measures objects in binary or threshold images. Once the image has been segmented we can obtain various information regarding particle size and numbers. By using ImageJ software we can also perform a set of measurement on a selected object (the brain tumor showed in Fig. 1.a.). The Integrated Density represents the sum of the values of the pixels in the selection, being equivalent to the product of Area and Mean gray value. Median (29) exhibits the middle value of the pixels in the selected brain tumor. The Feret s diameter (Caliper length = cm) is the longest distance between any two points along the selection boundary. The measurement results are presented in calibrated units (Table 1). A fundamental task in many statistical analyses is to characterize the location and variability of data set. Skewness is a parameter that describes the asymmetry of a PDF (Probability Density Function) while Kurtosis is a parameter that depicts the shape (the degree of peakedness - broad or narrow) of a PDF. For univariate data: X1,X 2,..., X n, the formula for Skewness and the Kurtosis are defines as: skew = n i= 1 ( n 1) s 3 3 _ X i X (4) and kurt = n i= 1 ( n 1) s 4 4 _ X i X where _ X is the sample mean, s is the standard deviation and n is the number of data points. The Skewness for a normal distribution is zero and the kurtosis for a standard normal distribution is three. This statistical measure was used to describe the distribution of observed data around the mean. Positive values for the Skewness (2.257) show that data are skewed right. Positive Kurtosis (6.511) indicates a peaked distribution. Table 1 - The measurement results of a brain tumor Area (cm 2 ) Min/ Max 4/203 Skewness StdDev Mean gray value Kurtosis Circularity Modal gray value 13 Perimeter (cm) Integrated Density Median 29 Feret s diameter (cm) Profile Plot displays a two-dimensional graph of the intensities of pixels along a line (x-axis or y axis) within the X-ray images (Fig.6. a,b). For plotting the S1 X-ray CT of abnormal brain I have been used the figure 1.a. High peaks depict low density tissues (lack of tissue) while boundary valleys with lowest gray values show calcified tissue in Profile Plots on x-axis and y-axis. (5) 406

7 In order to acquire the power spectrum as a function of frequency we have been applied the Fast Fourier Transform (FFT) analysis by using the histogram values (Fig. 7.a,b) or profile plot values (Fig. 8.a,b). Figure 6 - a,b - ImageJ Profile Plots of S1 X-ray CT abnormal brain scan a) on x- axis: = 5 cm, = 8.67 cm b) on y-axis: = 2.67 cm, y = 5.5 cm x1 x 2 y1 2 Figure 7 - a,b - OriginPro FFT analyze of (a) an X-ray CT normal brain scan and (b) S1- X-ray CT abnormal brain scan. The plots were performed by using the histogram values. Figure 8 - a,b - OriginPro FFT analyze of S1- X-ray CT abnormal brain scan on (a) x- axis and (b) y-axis. The plots were performed by using profile plot values of S1 X-ray CT abnormal brain scan Contour Plot is useful for delineating organ boundaries in images. The X-ray CT image of the abnormal brain can also be plotted using a graph template that includes X and Y projections. While the X-ray CT scan show an abnormal growth tissue (x = and y = 925.0), the Contour Plot (Fig. 9.a.) reveals additionally data about other brain tissue damages in both hemispheres. 3D Color Surface Map displays a three-dimensional graph of the intensities of pixels in a gray scale or pseudo color image (Fig.9.b). 407

8 Figure 9 - a,b OriginPro Profile Contour Plot 3D Color Surface Map of S2 X-ray CT abnormal brain scan 4 CONCLUSIONS Image enhancement technique allows the increasing of the signal-to-noise ratio and accentuates image features by modifying the colors or intensities of X-ray CT brain image. The X-rays penetrate the tissues differently depending on the type of tissue. The solid tissue, such as bone, appears white and the air appears black. Image processing of X-ray CT scans displayed the characteristic pattern of an abnormal brain showing calcified and lack tissues or asymmetric perfusion in both hemispheres correlated with the neurological disease. Image analysis with OriginPro 7.5 and ImageJ programs revealed Histograms, Profile Plots, Power Spectra and measurements on a brain tumor with a Feret s Diameter of cm. REFERENCES [1] Barrett, H.H.; Swindell, W.: Radiological Imaging, The Theory of Image Formation, Detection and Processing, Vol. I, Academic Press, New York, USA, , (1981) [2] Bistriceanu, E.G.: Principiile Matematice şi Fizice ale Tomografiei Computerizate, Editura Matrix ROM, Bucureşti, 17-27, (1996) [3] Duliu, O.G.: Computer axial tomography in geosciences: an overview, ELSEVIER, Earth-Science Reviews 48, , (1999) [4] George, M.S.; Ring, H.A.; Costa, D.C.; Ell, P.J.; Kouris, K.; Jarritt, P.H.: Neuroactivation and Neuroimaging with SPET, Springer-Verlag, London, p. 8, (1991) [5] Kak, A.C.; Slaney, M.: Principles of Computerized Tomographic Imaging, The Institute of Electrical and electronics Engineers, Inc., New York, p. 119, (1999). [6] Webb, S.: The Physics of Medical Imaging - Medical Science Series, Institute of Physics Publishing Bristol (Great Britain) and Philadelphia (USA), , (1996) [7] ImageJ, Corel PHOTO-PAINT 12.0, Adobe Photoshop and OriginPro 7.5 software and Tutorials [8] Seleţchi, E.D.: Adjustment Filters and particle Analysis on X-ray CT scans - The 3 rd International Scientific Conference else, , (2007) 408