Image Evaluation PDF

Image Evaluation Image Display Image Quality © Copyright 2016 Pulse Radiology Education  Image Display 1. Pixel, Voxel 2. Matrix 3. Image Magni cation...

Image Evaluation Image Display Image Quality © Copyright 2016 Pulse Radiology Education  Image Display 1. Pixel, Voxel 2. Matrix 3. Image Magni cation 4. Display Field of View (dFOV) 5. Window level, Widow Width 6. Cine 7. Region of Interest (ROI) © Copyright 2016 Pulse Radiology Education fi  Digital Image Formation After attenuation has been measured by the detectors and raw data has been acquired, image reconstruction takes place to form the CT digital image. During this process is when voxels based on patient’s anatomy is divided into a matrix and displayed as a matrix of pixels. © Copyright 2016 Pulse Radiology Education  Digital Image Formation The image displayed, although 2D, corresponds to 3D sections of the patient’s anatomy. Each pixel on the CT image displays the average x-ray attenuation properties of the tissue in the corresponding voxel. Pixels are determined by the computer program and not by the dimensions of the x-ray beam. CT scan format consists of many cells, each assigned a number and displayed as an optical density or brightness level on the video monitor. © Copyright 2016 Pulse Radiology Education  Digital Image Formation The resulting image is a cross-section of anatomy with an assigned CT number based on brightness. Superimposed two-dimensional images as opposed to cross-sections have signi cant loss of information. © Copyright 2016 Pulse Radiology Education fi  Pixel A single picture element. Multiple pixels make up the image matrix. Each pixel in an image represents the CT number of a volume of tissue at that position in the slice. Each pixel’s grayness represents the tissue’s density. Pixel size is determined by: the used eld of view number of elements in the display matrix. Pixels do not have xed sizes. Often used to measure resolution. More pixels = sharper image © Copyright 2016 Pulse Radiology Education fi fi  Voxel A volume element. Voxel dimensions are de ned by the slice thickness and by the in-plane spatial resolution. Essentially a pixel + depth. Pixel intensity is proportional to signal intensity of the appropriate voxel. © Copyright 2016 Pulse Radiology Education fi  Matrix The image matrix is a grid of pixels that form the image. The image matrix affects: The apparent noise Resolution of the resulting image © Copyright 2016 Pulse Radiology Education  Matrix Resolution = reconstruction FOV / matrix. An increase in matrix will result in pixels being smaller. Will allow smaller details to be seen in the image. Smaller pixels = better resolution. © Copyright 2016 Pulse Radiology Education  Matrix Increase in matrix also increases the appearance of image noise. When an x-ray beam is generated, there is only a nite number of photons. Only a nite of photons get detected. That nite number of photons must distribute the signal across more pixels when the matrix increases. Decrease in overall signal per pixel will result in increased noise. Most commonly used matrix size is 512 x 512. Value for matrix is not typically changed. © Copyright 2016 Pulse Radiology Education fi fi fi  Magni cation Allows the viewer to enlarge part of the reconstructed image. Larger image allows for easier observation of anatomical and pathological structures. Magni cation does not affect: Spatial resolution Appearance of noise Simply enlarges individual pixels for display purposes. Zooming in © Copyright 2016 Pulse Radiology Education fi fi  Magni cation A post-processing technique. Applied to image data not raw data. Differs from retrospectively reconstructing the eld of view. Reconstruction FOV may affect spatial resolution and apparent noise level of the image. Resolution = dFOV / matrix Only method available for increasing the display size of anatomy if the raw data is no longer available. © Copyright 2016 Pulse Radiology Education fi fi  Display Field of View May also be called reconstruction eld of view. Selectable scan factor measured from the center of the patient to the most distant located edge of the patient. It shows what you set your localizer to Always equal to or smaller than the scan eld of view. © Copyright 2016 Pulse Radiology Education fi fi  Display Field of View Small or narrow FOV is like a zoom lens. Restricts the area of interest and displays the anatomy larger. © Copyright 2016 Pulse Radiology Education  Display Field of View Large FOV will make anatomy appear smaller. Noise is less noticeable Compromises spatial resolution Determines the size of an object that can be seen on the resulting image. In-plane spatial resolution = reconstruction FOV/number of pixels along each direction in the matrix Spatial resolution is optimized using small a dFOV. Choosing the right FOV requires balance between noise and resolution. © Copyright 2016 Pulse Radiology Education  Window Width and Window Level Window width and window level allow us to change the appearance of the image. Spreads a small range of CT numbers over a large range of grayscale values. Makes it easier to detect small changes in CT numbers. Windowing: The process where the CT number range of interest is spread over the full grey scale on the display system. Standard window Lung window Bone window © Copyright 2016 Pulse Radiology Education  Window Width Determines the range of CT numbers displayed on an image. Contrast Any CT number above this range is assigned a pixel level of 255. Appears white. Any CT number below this range is assigned a pixel level of 0 Appears black. Within the range of the window width, the CT numbers are equally distributed from black to white. © Copyright 2016 Pulse Radiology Education  Window Width Increase in window width: Less displayed contrast between tissues with similar densities. Used when viewing structures with a high degree of contrast Lung window Narrow window width (decrease): Increases contrast between tissues of similar density. Allows for contrast differentiation. © Copyright 2016 Pulse Radiology Education  Window Level Sets the center CT number displayed on the monitor. Brightness Generally equal to the CT number of the tissue of interest. © Copyright 2016 Pulse Radiology Education  Window Level and Window Level Example: Window width of 500 is chosen and a window level of 0 is chosen, tissues with a CT number of -250 to +250 will be displayed. Any tissue with a CT number higher than 250 will be displayed as white. Any tissue with a CT number lower than -250 will be displayed as black. If the window level is changed to 250, the tissues with CT numbers from 0 to +500 will be displayed. © Copyright 2016 Pulse Radiology Education  Windowing Windowing: The process where the CT number range of interest is spread over the full grey scale on the display system. Allows us to focus on certain details. Example: Chest CTs may be visualized in the mediastinum or standard window to check for soft tissue structures and contrast enhanced structures. Bone window may be utilized to visualize bony trabeculae and bony detail to check for osteogenic pathologies in the thoracic spine or rib cage. Lung window may be utilized to visualize bronchial and pulmonary structures in the lung. © Copyright 2016 Pulse Radiology Education  Windowing © Copyright 2016 Pulse Radiology Education  Cine A mode that can be set to acquire a series of rapidly recorded multiple images taken at sequential cycles of time. Displayed in a dynamic movie display format. © Copyright 2016 Pulse Radiology Education  Region of Interest Identi cation of a given area of May be used in various different scenarios: an image for numerical analysis. Quality assurance testing The area being scanned that Recording measurements is of particular importance. Bolus triggering Can also be called a volume Although not suggested, may be used to of interest. determine pathology from normal tissue based on recorded CT number. © Copyright 2016 Pulse Radiology Education fi  Region of Interest When measuring a region of interest: Measuring using the smallest possible area. Exception is when taking a volume of an organ. Measure away from borders and edges to avoid partial volume averaging. © Copyright 2016 Pulse Radiology Education  Region of Interest ROIs are used during daily quality assurance testing. Measurements are taken from a water phantom. Typical recorded values are: Mean (average of CT numbers) Standard Deviation © Copyright 2016 Pulse Radiology Education  Region of Interest © Copyright 2016 Pulse Radiology Education  Region of Interest Mean: Average CT number of the pixels in the region of interest. In a water phantom, the CT number for water should be 0 +/- 3. Standard Deviation: Measurement used to quantify the amount of variation of a set of data values. The standard deviation of the mean. System manufacturers specify the acceptable ranges and the imaging technique to use. Out or range values may indicate an increase in image noise. © Copyright 2016 Pulse Radiology Education  Image Quality 1. Spatial Resolution 2. Contrast Resolution 3. Temporal Resolution 4. Noise and Uniformity 5. CT Number (Houns eld Units) 6. Linearity 7. Quality Assurance and Accreditation © Copyright 2016 Pulse Radiology Education fi  Image Quality Things we focus on when To achieve this, we must performing a CT exam: understand the key Producing the best in uences on CT quality image we can get. parameter selection: Keeping patient dose to Contrast resolution as low as reasonably Spatial Resolution achievable. There is a tradeoff Temporal Resolution between maximizing Image Noise image quality and minimizing dose. Patient Dose © Copyright 2016 Pulse Radiology Education fl  Image Quality © Copyright 2016 Pulse Radiology Education  Spatial Resolution Measure of size of the smallest object that can be visualized in an image. Expressed as line pairs How well can you see the smallest object? Goal is to be able to see all structures clearly. Factors in uencing spatial resolution: A. Geometric Factors B. Controllable Factors: Focal spot size Slice thickness Detector aperture size reconstruction eld of Focal spot to patient view distance Image matrix Patient to detector distance Reconstruction lters © Copyright 2016 Pulse Radiology Education fl fi fi  Spatial Resolution © Copyright 2016 Pulse Radiology Education  Contrast Resolution Ability to differentiate between small differences in density. Density differences of less than 0.5% are detectable on CT. In comparison, conventional radiography can only differentiate no less than 10% density difference. kVp controls the contrast resolution on CT images. © Copyright 2016 Pulse Radiology Education  Temporal Resolution Precision of a measurement with respect to time. How fast can the scanner acquire data? There is often a relation between temporal resolution and spatial resolution. At one point in history, a CT scanner was not capable of imaging faster than the motion of a patient. Spatial resolution was compromised. Technological advancements have made it possible to acquire a scan in a very short amount of time. © Copyright 2016 Pulse Radiology Education  Temporal Resolution As acquisition time became faster: Motion artifacts were no longer an issue Decreased scan times Shorter breath holds Newer applications Imaging of moving blood Imaging of the beating heart without motion artifact © Copyright 2016 Pulse Radiology Education  Noise A uctuation in CT number within something that should be uniform. Makes the image look grainy Becomes dif cult to notice subtle contrast differences or ne detail. Caused by undesirable electronic interferences. There is a direct correlation between noise level and number of photons used to create the image. © Copyright 2016 Pulse Radiology Education fi fl fi  Noise The less photons in the x-ray beam, the Easiest way to increase number noisier or grainier the image will look. of photons is by increasing To reduce noise, increase the tube current or scan time number of photons. (mAs). There are no parameters on the scanner that allow a direct decrease in the Disadvantage: Increased amount of noise. patient dose. Increasing patient dose should only be considered if the resulting image from not increase dose would be non- diagnostic. Only do this if absolutely necessary to produce a diagnostic image. © Copyright 2016 Pulse Radiology Education  Noise © Copyright 2016 Pulse Radiology Education  Uniformity During quality assurance testing, a CT number is recorded off the water phantom. This value should be 0 +/- 3. Based off the mean of the ROI The CT number in a homogenous object (water phantom), should be the same over various regions. Differences between the center and the periphery is largely due to then beam hardening effect. © Copyright 2016 Pulse Radiology Education  Uniformity Consistency of the CT numbers of an image of a homogenous material across the scan eld. Calculated as the difference in the mean CT number in the center from that at the periphery of the image. Routinely done as part of the daily quality control. Should always follow the prede ned manufacturer’s procedures for testing uniformity. Good uniformity is de ned as: for a uniform phantom, CT number measurements should not change with the location of the selected ROI or with the phantom position relative to the iso-center of the scanner. Variation in uniformity can indicate increased noise. © Copyright 2016 Pulse Radiology Education fi fi fi  CT Number Also known as Houns eld Units (HU). Named after Godfrey Houns eld In 1967 he produced the rst clinically useful CT scanner for brain imaging. fi fi fi  CT Number Relative comparison of x- ray attenuation of each voxel of tissue with an equal volume of water. Formula: The difference between the attenuation of the tissue relative to water divided by the attenuation of water multiplied by 1000. 1000 being the limit of the HU scale. © Copyright 2016 Pulse Radiology Education  CT Number Each pixel in the digital image is assigned a gray scale number. Gray scale numbers are assigned according to the CT number calculated by ltered back projection during image reconstruction. The digital image is made up of CT numbers from all the materials in the area of interest being scanned. © Copyright 2016 Pulse Radiology Education fi  CT Number CT number scale (houns eld scale) is based on the attenuation of water. On the HU scale: Water is always 0 HU Air is always -1000 HU Fat is approximately -80 HU Soft tissue ranges from 30-80 HU Any measurements above 100 usually indicates presence of calci cations. © Copyright 2016 Pulse Radiology Education fi fi  CT Number Characteristics of a CT number: High CT numbers (dense structures) are assigned lighter shades of gray. Lower CT numbers are assigned dark shades of gray. CT images contain CT numbers that range from -1000 (air) to +1000 (dense bone). Operator’s console can only display 256 shades of gray. Human eye can only distinguish about 20 shades of gray. © Copyright 2016 Pulse Radiology Education  CT Number Producing accurate CT numbers is important when comparing healthy tissue from disease pathology. Two aspects factor into CT number accuracy: Consistency: dictates if the same phantom is scanned with difference slice thicknesses at different times, or in the presence of other objects, the CT numbers of the reconstructed phantom will not be affected. Uniformity: dictates that for the uniform phantom, CT number measurements should not change with the location of the selected ROI or with the phantom position relative to the iso-center of the scanner. © Copyright 2016 Pulse Radiology Education  Linearity Linearity: property of a system, characterized by output that is directly proportional to the input. In CT, this means the amount to which the CT number of a material is directly proportional to the density of the material in houns eld units. Linear relationship between the calculated CT number and the linear attenuation coef cient of the object. © Copyright 2016 Pulse Radiology Education fi fi  Linearity The CT unit must be calibrated regularly to ensure accuracy of the numbers each time. Accuracy between the linear attenuation coef cient and the CT number can also be used to describe the performance of the CT scanner. Deviations from linearity should be < +/- 5 HU. © Copyright 2016 Pulse Radiology Education fi  Quality Assurance and Accreditation Procedure of performing speci ed tests or measurements on a periodic basis. Assures that a set level of quality, as speci ed by the system manufacturer has not been compromised. Important that daily quality assurance testing be performed. This is the only method of measuring, recording, and tracking daily quality assurance data that slow drifts in system performance will be noticed and xed. © Copyright 2016 Pulse Radiology Education fi fi fi  Quality Assurance and Accreditation Over the years, quality assurance measurements have simpli ed and minimized due to: Hardware stability has increased. CT system software has built in self-assessment procedures to assure system stability. Requirement for quality assurance tests will vary from scanner to scanner. Manufacturers provide recommendations for tests performed. Other weekly, monthly, and/or annual testing by be performed by the service engineer, or medical physicist. © Copyright 2016 Pulse Radiology Education fi  Quality Assurance and Accreditation Most frequently used standards for daily quality assurance assessment: CT Number Calibration Test CT Number Standard Deviation Test Performed on a water phantom. © Copyright 2016 Pulse Radiology Education  Quality Assurance and Accreditation In daily quality assurance testing, after the image is reconstructed, a region of interest is placed to select a region in the center of the phantom. Exact speci cations are provided by the system manufacturer. On some scanners, this may be an automated procedure. Within the region of interest, two parameters are evaluated: 1. Mean 2. Standard deviation of the CT numbers © Copyright 2016 Pulse Radiology Education fi  Quality Assurance and Accreditation Acceptable limit for CT Number Calibration Test -3 to +3 Ideally, CT number of water should be 0, but a variation of +/- 3 is allowed. Recalibration may be required if it falls outside the acceptable range. Service representative must be noti ed if it fails again. © Copyright 2016 Pulse Radiology Education fi  Quality Assurance and Accreditation Acceptable limit for CT Number Standard Deviation Test: Dependent upon scan parameters used. System manufacturers specify technique and acceptable range of values. If measured value is out of range: Indication of an increase in image noise. Could be caused by decreased in delivered dose. Increase in image noise in the detector-ampli er chain. Warrants noti cation of the service representative. © Copyright 2016 Pulse Radiology Education fi fi  Quality Assurance and Accreditation Phantom is moved into position so that the center of the water section is isocenter. Centered to the x-ray beam. Serial/axial scan is performed over the water phantom. 5 mm reconstructed slices Standard brain CT acquisition parameter. View central reconstructed slice. With scanner analysis software, draw a circular ROI in the center of the image. Diameter should be about 10% of the image width. Record CT number Repeat measurements at the periphery in 4 positions: 12 o’clock, 3 o’clock, 6 o’clock, 9 o’clock. © Copyright 2016 Pulse Radiology Education  Quality Assurance and Accreditation There are many other tests performed on a less frequent basis. May be performed by medical physicist, or service engineer. Tests include measurements of: Linearity of gray scale of image Spatial resolution as it relates to actual density of tissues measured Contrast resolution Radiation scatter and leakage Table accuracy Slice thickness accuracy Dose Fidelity of the video monitor Accuracy of spatial and lm output. measurements on image Other manufacturer speci c Consistency of CT tests. numbers across the image © Copyright 2016 Pulse Radiology Education fi fi  Quality Assurance and Accreditation A strong quality control program will make the process of achieving accreditation easier. Certain states require that certain facilities/departments be accredited by a nationally recognized accrediting body. Example: In New York State, accreditation by a nationally recognized accrediting body acceptable to the Department of Health is a requirement for the radiation permit issued to free-standing radiology facilities. New York State recognizes: American College of Radiology (ACR) Joint Commission on the Accreditation of Healthcare Organizations (JCAHO) Intersocietal Accreditation Commission (IAC) © Copyright 2016 Pulse Radiology Education  Quality Assurance and Accreditation Accreditation process consists of reviewing the facility’s quality control program. Quality assurance testing and recorded results Physicists reports Service reports State inspections The facility’s image quality may also be reviewed. Checks to make sure facility is operating at safe levels of radiation dose. Maintains a standard to prevent unnecessary exposure to the public. © Copyright 2016 Pulse Radiology Education  Quality Assurance and Accreditation Other reasons a facility may wish to pursue accreditation: Marketing advantage Reimbursement purposes. Medicare Improvements for Patients and Providers Act of 2008 (MIPPA) All physicians who bill under Part B fee schedule must be accredited by Jan 1, 2012 in order to receive technical component payments from CMS for advanced imaging. No accreditation = no payment for performing a CT. Other insurance carriers have followed this model. This does not apply to hospitals. Hospitals are accredited by JCAHO. © Copyright 2016 Pulse Radiology Education  Test Questions 1. What is the expected result of a CT Number Calibration Test performed on a water phantom? A. 0 +/- 3 * B. 0 +/- 5 C. 1000 +/- 5 D. 100 +/- 3 2. CT numbers ______ would appear black if a window width of 400 and a window level of -100 was used. A. above +200 B. below -300 C. below -200 D. above +100 * 3. Which of the following describes how grainy or speckled a CT image appears? A. Resolution B. Linearity C. Noise * D. Contrast 4. CT Number Calibration Test should be performed: A. Weekly B. Monthly C. Annually D. Daily * © Copyright 2016 Pulse Radiology Education  Test Questions 5. If a calci cation with CT number of +300 appears pure white on the image, what could be the window width and window level selected? 1. window level = 0, window width = 500 2. window level = +300, window width = 500 3. window level = +100, window width = 1000 A. 1 only * B. 2 only C. 3 only D. 2 & 3 only 6. Which of the following is TRUE about spatial resolution of an image? 1. Resolution is dependent on matrix size 2. Resolution is dependent on reconstruction FOV 3. resolution is dependent on scanned FOV A. 1 only B. 2 only C. 1 & 2 only * D. 1, 2 & 3 7. If a reconstruction FOV of 32 cm and a 512 x 512 matrix are used, the pixel size is ____ mm A. 0.35 x 0.35 B. 0.625 x 0.625 * C. 1.6 x 1.6 D. 3.25 x 3.25 © Copyright 2016 Pulse Radiology Education fi  Test Questions 8. Which of the following will result in an increased voxel size? 1. Decreased reconstruction FOV 2. Increased slice thickness 3. Increased matrix A. 1 only B. 2 only * C. 2 & 3 only D. 1, 2 & 3 9. Which of the following occurs when reconstruction FOV is decreased? 1. Spatial resolution is increased 2. Pixel size is increased 3. Noise in the image increases A. 1 only B. 2 only C. 1 & 3 only * D. 1, 2 & 3 10.There are a total of _____ pixels in a 1024 x 1024 matrix A. 2048 B. 524,288 C. More than 1 million * D. More than 2 million © Copyright 2016 Pulse Radiology Education  Test Questions 11. The smallest object that can be resolved if a reconstruction FOV of 24 cm and a 512 x 512 matrix are used is ___ A. 0.47 * B. 1.00 C. 2.13 D. 3.57 12.Which of the following parameters affects the contrast of a CT image? A. Reconstruction interval B. mAs C. kVp * D. CT number 13.Which of the following would have the highest Houns eld unit value? A. Blood * B. Water C. Fat D. Air 14.The process of adjusting the image so that the CT number for the tissue of interest becomes most pronounced is known as: A. Windowing B. Calibration C. Quality assurance D. Linearity © Copyright 2016 Pulse Radiology Education fi  Test Questions 15. Which of the following would be used to better visualize low contrast resolution? 1. Decrease mAs 2. Sharp reconstruction lter 3. Narrow window width A. 1 only B. 2 only * C. 2 & 3 only D. 1, 2 & 3 16.Which of the following is TRUE? 1. Magni cation requires the raw data 2. Increasing magni cation increases resolution 3. Magni cation decreases the size of the displayed anatomy A. 1 & 2 B. 2 & 3 C. None of the above * D. 1, 2 & 3 17.If a thin slice is used, which of the following would reduce the apparent image noise? A. Reduced reconstruction eld of view B. Increased matrix C. Increased mAs D. Increased kVp © Copyright 2016 Pulse Radiology Education fi fi fi fi fi  Test Questions 18. Increasing which of the following parameters would yield the best spatial resolution? 1. Field of view 2. Matrix 3. Slice Thickness A. 1 only B. 2 only C. 1 & 2 only * D. 1, 2 & 3 19.What window should a chest CT be viewed in? 1. Standard 2. Lung 3. Bone A. 1 & 2 B. 2 only C. 2 & 3 D. 1, 2 & 3 * 20.The quality of acquisition time is known as: A. Temporal resolution * B. Spatial Resolution C. Uniformity D. Linearity © Copyright 2016 Pulse Radiology Education  Resources Medical Imaging Consultants, Inc: CT Registry Review 4th Edition (2010) Sprawls: Computed Tomography Image Formation: http://www.sprawls.org/resources/CTIMG/module.htm#24 Sprawls: Computed Tomography Image Quality and Optimization: http://www.sprawls.org/resources/CTIQDM/ #13 NY Department of Health CT Quality Assurance: https://www.health.ny.gov/regulations/recently_adopted/docs/ 2016-01-20_computed_tomography.pdf Zeman M.D. R.K.; Why Seek Accreditation of Your CT Program?: http://www.imagewisely.org/imaging- modalities/computed-tomography/imaging-physicians/articles/why-become-accredited-by-the-ct-accreditation- program Radiology T-I-P: www.radiology-tip.com/ Physics Central: http://www.physicscentral.com/explore/action/scans.cfm Principals and Basics of Computed Tomography: http://www.slideshare.net/DEEPAK01/ct-its-basic-physics CT Number Accuracy and Noise: https://spie.org/samples/PM259.pdf W6 CT Uniformity: http://www.dap.org/CmsFiles/File/Safety%20Code%20HC35/Test%20Protocol%20Files/ W6.pdf “Pixel” Graphical Illustration: http://images.slideplayer.com/15/4563816/slides/slide_5.jpg “Voxel” Graphical Illustration: http://www.yoshida-net.co.jp/en/products/x-raysystems/ct/ necube/img/ ct_hako_en.gif © Copyright 2016 Pulse Radiology Education fi  Resources “Width and Level” Graphical Illustration: http://www.radiantviewer.com/dicom-viewer-manual/ radiant_dicom_viewer_adjust_window_mouse_moves.png” “Cine” Graphical Illustration: https://www.researchgate.net/pro le/Marcus_Carlsson2/publication/41532497/ gure/ g2/AS:277023818829834@1443059033621/Figure-2-Multidetector-contrast-enhanced-cine-CT- images-acquired-at-a-end-diastolic.png “QA” Graphical Illustration: https://www.ceessentials.net/images/qualitycontrol/image05.jpg “Phantom scan” Graphical Illustration: https://www.ceessentials.net/images/qualitycontrol/image01.jpg “Bolus” Graphical Illustration: https://www.med-ed.virginia.edu/courses/rad/CTPA/images/smartprep.jpg “ROI” Graphical Illustration: https://www.researchgate.net/ gure/262842660_ g1_Fig-1-Region-of-interest- placement-in-target-vessels-a-Thin-section-axial-CT-image “ROI Liver” Graphical Illustration: http://www.santesoft.com/win/sante_ct_viewer/img/roi_tool.png “HU Scale” Graphical Illustration: http://crashingpatient.com/wp-content/images/part1/houns eld2.jpg “HU Chart” Graphical Illustration: https://www.ceessentials.net/images/qualitycontrol/image03.jpg “Godfrey Houns eld” Graphical Illustration: http://www.nobelprize.org/nobel_prizes/medicine/laureates/1979/ houns eld.jpg “Brain window” Graphical Illustration: http://www.aspectsinstroke.com/typeroom/assets/uploads/windowing- levels.jpg “Brain windowing” Graphical Illustration: http://image.slidesharecdn.com/approachtoheadct-130418093230- phpapp01/95/approach-to-head-ct-12-638.jpg?cb=1366277727 © Copyright 2016 Pulse Radiology Education fi fi fi fi fi fi fi fi  Resources “Neuro” Graphical Illustration: http://radiology.ucla.edu/images/specialties/header-neuro.jpg “Futuristic Radiology Display” Graphical Illustration: http://il9.picdn.net/shutterstock/videos/ 11118257/thumb/2.jpg “Future healthcare” Graphical Illustration: http://il3.picdn.net/shutterstock/videos/5956985/ thumb/4.jpg “Voxels, Pixels” Graphical Illustration: http://clinicalgate.com/wp-content/uploads/2015/06/ c00022_f022-006-9781455753048.jpg “Lung windowing” Graphical Illustration: http://www.people.vcu.edu/~mhcrosthwait/PETW/ images/wwwlexamples.jpg Chest windows”Graphical Illustration: http://www.svuhradiology.ie/wp-content/uploads/ 2015/04/CTwindows.jpg “Linearity” Graphical Illustration: http://slideplayer.com/slide/9118416/ “IAC” Graphical Illustration: http://www.bayareaentspecialists.com/images/news/ICACTL.jpg “JACHO” Graphical Illustration: http://www.bayareaentspecialists.com/images/news/ ICACTL.jpg “ACR” Graphical Illustration: http://www.bayareaentspecialists.com/images/news/ICACTL.jpg © Copyright 2016 Pulse Radiology Education