Computers in NMT
How to calculate percent energy resolution?
% ER= FWHM/energy of radionuclide in Kev x 100%
calculation of number of pixel locations
(#x locations)(#y locations)
How many bytes are in a Gigabyte
1 billion
What is the matrix depth for word mode, counts per pixel?
2^16-1=65,535 counts per pixel High counts studies should be recorded in this mode
What is the number of counts per pixel for byte mode?
2^n-1= 2^8-1=255 counts per pixel n= number of bits
What size matrix is used for a planar high count rate study that has the best resolution?
512x512
What matrix size is used for most SPECT imaging?
64x64
What is the gamma photopeak of Cs-137?
662 KeV
What is the range the percent energy resolution must fall into?
8% to 12%
How does matrix size correspond to memory?
A 64x64 matrix can hold 4096 pixels. Total memory is the number of columns times the number of rows
Explain what study uses the 128x256 matrix and why?
A whole body, and we need only half the columns because the width doesn't have to occupy the whole FOV, but the length needs to larger to accommodate the length of the body
Why should low count studies be saved in byte
Bc it takes less counts per pixel to fill so the focal spots will show up as brighter.
The smaller the FWHM the _________________ the resolution.
Better
What do modern computers use to process information?
Binary digits that act as on and off switches- digital memory uses binary digits
What is the difference between a Bit, Byte, and Word?
Bit- single block of information Byte- 8 bits- represent a single character Word- 2 bytes (16 bits)
Where does imaging processing and display occure
CPU
What is one possible undesirable side effect from image smoothing?
Can add too many counts and miss a cold spot
How does image smoothing occure
Can smooth over by adding counts in.
How is matrix selected?
Count rate, memory and how the information must be stored- byte or word
Voxel
Describes the depth of the matrix
An example of adding images together?
Dynamic flow study images collected at one second each can be added, three one second images can be added to make one three second image. Sometimes done on renal imaging, to improve the resolution when drawing the ROI
Computer in Nuclear Medicine are used for the following: A. Acquire- Input B. Storage C. Processing of data- central processing unit D. Output E. All of the above
E. All of the Above
How is one count added to the storage location in the computer
Each Z pulse that is accepted by the PHA adds one count which is stored with x and y coordinates.
T/F High count studies depth of storage is in byte mode.
False- counts could be lost because the depth isn't there to store the counts.
What mode is static images stored in?
Frame mode, information is placed on the matrix without regard to time
What kind of studies are 256x256 matrix used for?
High count rate, the resolution is much better because the pixels can have their own coordinates.
What is frame rate
How long are counts being collected per frame.
What factor determines the intensity of the image?
Image density per pixel, which is determined by the number of gamma counts per pixel
What must you submit when submitting a histogram?
Images showing the organ with drawn ROI used to verify correct placement
When can image subtraction be done?
In dual isotope studies, parathyroid. Can subtract the thyroid from the parathyroid images to see the parathyroid study by using the windows for Tc and I-123 to separate the isotopes
How is the information placed for static images(4)?
Info is placed on one matrix over a period of time, example bone. Matrix size- 128x128 Info is superimposed on the same matrix over time. Stored in computer in frame mode
What is rollover?
Information is deleted or the computer stops collecting counts when it reaches the count limit per pixel when pixel counts are exceeded.
What happens to the information if the sequence is too long
Information is lost- temporal resolution is lost. May miss the flow of the RP in a renal
What mode are gated images stored in?
List mode, how are they collected per completed heart beat, R to R interval. Time per frame= RR interval/# of frames
How are dynamic images stored?
List or histogram mode (counts vs time)- they are stored in order so the images can be played back in order
When would you use a 128x128 matrix?
Medium count rate study like a foot or kidney
Is resolution important on a flow study like renal?
No, important you see the function for these kind of studies.
What does normalization help prevent?
Over estimation or underestimation of background counts based on image size
Image manipulation- what is background subtraction?
Setting a count threshold and having it count to the lowest intensity setting
The larger the matrix, the __________ the pixel and the ___________ the image resolution
Smaller the pixel size Better the image resolution
What kind of resolution is important with flow studies?
Temporal resolution- don't make the time/ frame to long because you could miss the flow
What changes the analog information into digital form?
The ADC- analog to digital converter
What is image algebra?
The addition, subtraction multiplication and division of images.
Why is a small matrix good for a low count study?
The count will appear bigger because more pixels will occupy the same box, have the same coordinates. The counts can be seen better if they are limited to a small matrix
The number of columns is which number in a matrix?
The first
What do the modes in byte or word refer to?
The matrix depth
What defines a matrix size?
The number of columns and the number of rows
What defines the second number in a matrix?
The number of rows.
Calculation of number of images displayed
Total pixels/ Total pixels per frame
T/F Our cameras present the information in analog form. It is converted into digital form before the computer can use it.
True
Why would you want to use a smaller matrix size?
When you are conducting a low counts study like a planar flow study- renal or 3 phase bone.
How does pixel size change with zoom?
Zoom decreases pixel size Ex. Pixel size of a 64x64 matrix on a zoom is equivalent of the pixel size on a 128x128 matrix. As matrix size increases, pixel size decreases
How is image quantitation achieved?
a ROI is drawn A histogram is generated- generates time-x/ Activity counts-y curve
As you lose energy resolution the % energy resolution gets __________?
bigger
How can pixel size be determined when using Zoom?
distance between sources/ # of pixels between sources Diameter of the crystal/ number of pixels(number of columns) zoom factor is usually one
When converting from byte to word?
divide by 2
How are these images viewed?
in sine, bc the memory keeps tract in time sequence and they can be compressed
What is normalization and give an example?
it standardizes count information by keeping the count rate the same in variations of R to R intervals An example is a Muga Scan
The larger the matrix the _______________ the number of pixels available. The _______________each pixel.
larger smaller
What kind of source is used to evaluate energy resolution?
long lived, monoenergetic source Cs-137
Zoom, the images gets ___________ while the pixels get ___________.
magnified smaller
What is the pixel size based on?
matrix size
What does zoom not change?
memory because the same number of pixels are being used
When converting from word to byte, what do you do?
multiply by 2
How is the number of images displayed determined?
number of images displayed= total pixels on display screen/ total pixels per frame Matrix size is a function of how many images can be displayed
What is the zoom factor usually
one
What is the address location of a count on the matrix?
pixel
How is pixel size determined if the matrix size and camera FOV are known?
pixel size= FOV size/ column number of matrix 0.5 cm= 32 cm/ 64
The wider the curve the ___________ the energy resolution is?
poorer
What environmental factors must be maintained?
temperature- no quick changes humidity- keep around 50% Power- a dedicated source cleanliness- dust and dirt eliminated
How many components are there to a matrix
two- x and y component
