Topic 12: Image acquisition Flashcards
What is the effect of using more pixels ?
better spatial sampling (more detail)
but fewer counts per pixel therefore more noise.
How should you choose a choice of matrix?
How much noise we can tolerate.
How much detail we need to capture. - sampling theory states that pixel size should be at most 1/2 the object size.
How large would a pixel be if we were using a 400mm x 400mm detector and a 64 x 64 matrix?
If we were using an object 4mm in size, would our pixel size be large enough to display our object?
Pixel size = 400/64 = 6.25mm
Pixel size is therefore not adequate to image a 4mm object it would need to be at least 2mm.
Using too large a pixel size?
Overestimates the size of the object and underestimates the number of counts because of averaging of the high intensity and low intensity components.
What is the advantage of using digital zoom
allows us to use a smaller amount of the detector field of view.
reduce the amount of empty space when imaging small objects such as the thyroid.
How much noise would we have in 100 counts?
STD = 10 therefore noise = 10/100 = 10%
what is a way of reducing noise by acquisition time?
increasing acquisition time however that would increase latency.
How can you affect activity to affect noise?
What is the problem with this?
increase the acitivity.
However this strategy assumes a linear relationship between detector response and imaged activitiy. This exists up to a point, until pulse pile up count losses occur.
Name the simplest noise reducing filter that can be used?
Averaging filter.
The problem with smoothing filter?
degrades spatial resolution
What is the 9-point average filter? and what is a better alternative for digital image filtering to reduce noise?
average of a nine neighbouring pixels are taken and placed in the middle pixel. Done for all pixels in the image.
However this gives each neighbouring pixel equal value. A better solution is a nine-point smooth kernal which gives greater weight to central pixels.
What is the frame mode acquisition mode?
image accumulated using pre-defined parameters.
set pixel size, energy, matrix size, and pixel depth.
If we can acquire a frame we can acquire it statically or dynamically. Explain the difference and why we would use each one?
In static: For many types of radiopharmaceuticals after the initial take up period, the distribution is stable over time.
we produce one image over a given time period and is used to image a stable distribution of a radiopharmaceutical.
Dynamic imaging: For some imaging we want to capture the change in activity distribution over time to examine a physiological process.
This requires ‘dynamic’ imaging, where we produce many images, possibly with different durations to capture the process.
What is gated imaging and why do we need it.
We need time sensitive scanners to image fast moving things like a single phase of a cardiac cycle. So if we link it up to an ecg trace and spilt this trace into 8 or up to 24 bins.
An advantage of dynamic or gated inmaging : is we can visualise temporal changes in physiology. What is a way of quantifying these?
Region of interest analysis of time activitiy curves.
compare frame and list mode.
