Nuclear Medicine Processing and Acquisition

1
Nuclear Medicine Processing and Acquisition

• By Group One
• Gustavo Zambrano, Daniela Growgasapatawhacnamecall
  it, Chris Lequirica, Macarena Ayala, Jobe
  idicula,Alonzo Ventura

2
Objectives

• Describe image display and processing
• Explain what is Logarithmic and Exponential
• Describe Shades of Color
• Explain what is Cinematic Display
• Describe Image Quantitation

3
Objective

• Explain what is Curves
• Explain Normalization
• Example of Normalization
• Emission Computed Tomography (ECU)
• Explain Brain Death Processing and Acquisition

4
Objective

• Explain and teach Renal Processing and
  Acquisition
• Describe Myocardial Perfusion Processing and
  Acquisition
• Explain Combine Gastric Emptying
• Conclusion
• Resources
• Questions and Answers

5
IMAGE DISPLAY AND PROCESSING

• Gamma-ray scintillation events that occur at
  specific detection location are stored as digital
  images by accumulating gamma ray counts.

6
IMAGE DISPLAY AND PROCESSING

• These counts are stored in pixels which are then
  stored in byte or word. An image makes up an
  array of pixels.
• Each pixel is assigned a gray scale number based
  on the number of counts it has.

7
IMAGE DISPLAY AND PROCESSING

• The relationship between the number of counts and
  display intensity is usually linear.
• This provides a uniform shading between all count
  levels.
• Most computers allow a logarithmic or exponential
  relationship between pixel count and density

8
Logarithmic and Exponential

• Logarithmic assigns more gray levels to lower
  count pixels and compresses the shades of gray to
  the higher count pixels, enhancing differences in
  low count densities
• Exponential suppresses the number of gray scales
  assigned to low count pixels while expanding it
  for high count pixels. This helps remove
  background
• Background subtraction selects a count threshold
  that is set to the lowest intensity and reassigns
  intensities between the threshold and the maximum
  pixel count

9
Shades of Color

• The use of color is also use to provide image
  enhancement.

10
Shades of Color

• A typical color table is selected to enhance
  differences between pixel count densities and to
  provide some visual background erase.
• The most effective color tables are those that
  have gradual and continuous shades of color.

11
Image Algebra

• The simplest processing operations are those that
  represent mathematical operations of add,
  subtract, multiply and divide.
• Example A dynamic flow study originally acquired
  1 frame/sec can be reformatted through the
  addition of three images into one to create a
  dynamic set with each image representing 3
  seconds.

12
Image Smoothing

• Is performed to reduce noise from the random
  effects of radionuclide counting. The simplest
  technique is to average the counts of a given
  pixel with that of its eight surrounding
  neighbors and replace the center pixel with a new
  value. The most common type of image smoothing
  is the nine point smooth which is a filtered
  technique used to modify a specific pixel value
  according to the values of its neighbors.

13
Image Smoothing

• Smoothing creates a nicer more aesthetic image
  but it results in a slight loss of resolution and
  detail is slightly blurred. Filters with
  negative values around the edges can enhance or
  sharpen the edges of organs.

14
Image Smoothing

• Temporal filtering is used to filter dynamic
  images. It performs weighted averaging of an
  image with those that occur just before and after
  a dynamic image sequence. This type of filtering
  removes noise without a loss of spatial
  resolution

15
Frequency Space and Filtering

• The representation of images can be described as
  a group of frequencies. Image information can be
  represented in frequency space by graphically
  plotting the frequency on the horizontal axis and
  the wave amplitude of the vertical axis. In
  frequency space filtering there is always a
  trade-off or balance between reducing noise and
  degrading resolution.
• Frequency space can be thought as analogous to a
  piano keyboard. The low key create long
  frequency tones while the upper keys create the
  high frequency tones. Any object can be
  represented by a group of sine or cosine waves.
  And any image can be changed by changing a single
  frequency.
• When images are converted to frequency space,
  objects or organs are represented by low or
  medium frequencies while high frequencies are
  seen as noise.

16
Frequency Space and Filtering

• Cutoff frequency is a common parameter used to
  generate and characterize filter shapes and may
  be measured in cycles/pixel, cycles/mm or just
  millimeter. Commonly used filters are the ramp,
  von Hann or Hanning, Butterworth, Weiner and Metz
  filters. All these filters have the same basic
  purpose to increase the amplitude of the objects
  frequencies and reduce the amplitude of the high
  frequencies.
• Low pass filters allow low frequencies. These
  produce a higher resolution image but with more
  noise. High pass allow high frequencies and have
  a smoother effect but reduce resolution.

17
Cinematic Display

• Dynamic sequence of images may be displayed as a
  continuous- loop movie known as cinematic
  display.

18
Cinematic Display

• The images to be displayed are formatted into an
  area memory known as buffer so that information
  can be retrieved quickly.

19
Image Quantitation

• Counts in a particular area can be extracted from
  the image by defining a region of interest (ROI).
  
• An ROI is defined on the displayed image using a
  mouse, trackball, light pen or joystick.

20
Image Quantitation

• ROIs can be rectangular, elliptical, or
  irregular shaped.
• The area defined in an ROI should be
  physiologically meaningful.
• Different information can be extracted from
  images, depending on how the ROIs are drawn.

21
Image Quantitation

• Some nuclear medicine computers allow ROIs to be
  manipulated just as images can be manipulated.
  It might be desirable to add or subtract regions.
  

22
Curves

• Physiological information from dynamic studies
  might be appreciated more easily by generating
  time- activity curves. This curve displays are
  widely assigned to a variety of useful clinical
  applications.

23
Curves

• Curve provide useful information in evaluating
  the accumulation and washout of
  radiopharmaceutical from the kidneys, changes in
  left ventricular volume on gated studies and
  changes in radionuclide distribution in
  gastrointestinal studies.

24
Curves

• You can modify the curve appearance by making it
  a continuous line versus dots at different data
  points.

25
Normalization

• Concept in nuclear that implies that a
  measurement has been brought to a standard. For
  instance, 2 images with different maximum counts
  may have their lowest intensities normalized if
  the image with the lowest maximum count is
  multiplied so that the max count matches the max
  count in the second image. The 2 images would
  therefore be displayed with the same max
  intensity. Normalization is commonly applied to
  ROIs and curves.

26
Example of Normalization

• I123IBZM activity distribution during scan 1
  (baseline scan) and scan 2 (B,scan obtained at
  the end of AMPT administration, 1 g four times a
  day for 2 days) in 29 year-old male healthy
  volunteer. Both images were normalized to the
  I123 IBZM infusion rate, decay corrected for the
  beginning of the infusion and color coded.
  Striatal to occipital activity ratio was
  increased in the post-AMPT scan compared to the
  baseline scan.

27
Emission Computed Tomography (ECU)

• ECU is a general term describing the
  reconstruction of 3D image volumes that are
  deprived from one or two techniques (SPECT or
  PET)

28
Brain Death Processing and Acquisition
29
Brain Death Processing and Acquisition
30
Renal Processing and Acquisition
31
Renal Processing and Acquisition
32
Renal Processing and Acquisition
33
Renal Processing and Acquisition
34
Myocardial Perfusion Processing and Acquisition
35
Myocardial Perfusion Processing and Acquisition

• Processing
• Ensure that the study has an accession number
  using Patient Rename.
• Select the appropriate rest and stress SPECT
  images.
• Select QGS/QPS under Cardiac Apps
• Click Start
• Set the limits and orientation on VLA/HLA
  images for both stress and rest (pictured below)

36
Myocardial Perfusion Processing and Acquisition
37
Myocardial Perfusion Processing and Acquisition

• Align and match the slices.
• Change the combo box at the
• bottom to Current in order to move
• only one row.
• Change it back to All to move all of
• a projections slices at once.
• Right and Left Click on a row of images to move
  the
• slices.
• This is pictured above and to the left.
• Add annotation using the Overlay Annotation
  button,
• then clicking where you want to type your
  annotation (such as initials, or Stress motion
• corrected). Overlay annotations will be saved
  with the results.
• Click the intensity tab. Change the intensity
  combo box to All, and the max value to 140 and
• the minimum value to 10.
• You may need to type these numbers in to have
  the correct intensities.
• This is pictured to the right.

38
Myocardial Perfusion Processing and Acquisition

• This is pictured to the right.
• CTRL-Left click to remove the red box on the
  screen.
• Perform a screen capture AND print a film until
  directed otherwise.
• Name the screen capture Slices 1.
• Adjust the rows to show the rest of the slices,
  and repeat filming.
• icon. Name this screen capture Slices 2.
• Adjust the rows so that they are at the
  beginning again.
• Click the Process
• Click Quantitative Gated SPECT. It will
  automatically process.

39
Myocardial Perfusion and Processing and
Acquisition

• Print the Results and Views pages as
• you did the slices.
• o Name these screen captures
• QGS Results and QGS
• Views, respectively.
• o Orient the 3D model of the
• heart in all views so that the
• apex is NOT at an oblique
• angle (as pictured to the
• right )
• Click Quantitative Perfusion SPECT.
• It will automatically process.
• Under File, click
• Save and Exit.

40
Myocardial Perfusion and Processing and
Acquisition

• Select the entire study, and click on ECToolbox
• under Cardiac Apps.
• Click Start.
• Click ECToolbox under the Process tab.
• With the setting page, make sure all fields are
• correct especially gender.
• Click OK.
• Set your regions for the Stress and Rest
  images,
• and click Polar Map (pictured below)
• Under File, click Save and Exit.

41
Myocardial Perfusion and Processing and
Acquisition

• If motion correction was performed, send the raw
  data
• and the CARDIAC SPECT and Card view Results
• from that first processing to the Inbox using
  QGS/QPS
• More detailed information on Motion Correction
  with
• QGS/QPS is found in the Motion Correction SOP.
• Send the raw data and the CARDIAC SPECT ,
• Card view Results, and QGS/QPS Results to the
• Inbox using QGS/QPS.
• Then select all the raw data, the second (or
  third,
• if motion correction was performed) Card view
• Results, the ECTb_Results, and all snapshots
• and send it to the Inbox under ECToolbox.
• Ensure that you have the stress dose slip and
  stress
• history attached to the requisition
• Ensure the stress history has
• o Time and Heart rate for treadmill studies
• o Dosage of adenosine
• o Full sheet for dobutamine
• Ensure that this study is properly billed in
  IDX.

42
Combine Gastric Emptying
43
Combine Gastric Emptying
44
Combine Gastric Emptying
45
Conclusion

• Now we know how Gamma-ray scintillation events
  that occur at specific detection location are
  stored as digital images by accumulating gamma
  ray counts and how counts are stored in pixel,
  word, and byte
• The important of logarithmic, exponential, and
  background play a visual pixel role
• Now we know how important shades of color can
  distinguish in multiple level
• How the curve help us in visualization in
  multiple information in a single image
• The differences the visualization characteristics
  of dynamic, image quantization, normalization.
• How we explored the processing and Acquisition
  for the brain death, renal, myocardial perfusion,
  and gastric emptying

46
Resources

• surge.ods.org/xeleris/xeleris.htm