Imaging - Preprocessing and Histograms

0.0(0)
studied byStudied by 0 people
0.0(0)
full-widthCall Kai
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
GameKnowt Play
Card Sorting

1/51

encourage image

There's no tags or description

Looks like no tags are added yet.

Study Analytics
Name
Mastery
Learn
Test
Matching
Spaced

No study sessions yet.

52 Terms

1
New cards

What is preprocessing?

Manipulations made to the raw digital image data to compensate for physical flaws in image acquisition

Occurs before storage

2
New cards

What kinds of flaws does preprocessing correct?

Physical flaws related to:

1. Elements and circuitry of the image receptor

2. Physical elements and circuitry of the processor

3
New cards

What is point processing?

Adjusting the image brightness and contrast

Involves the creation of a histogram and application of a LUT

4
New cards

What is point processing also known as?

Grayscale processing

5
New cards

What is local processing?

Zooming and magnification (moreso post-processing)

6
New cards

What is geometric processing?

Changing the position or orientation of pixels (rotation)

7
New cards

On a histogram of a chest, where would the abdomen, mediastinum, lung tissue, and background be represented?

Abdomen: Left, brighter

Mediastinum: Middle

Lung: Right, darker

Background density is represented by spike to the right

<p>Abdomen: Left, brighter</p><p>Mediastinum: Middle</p><p>Lung: Right, darker</p><p>Background density is represented by spike to the right</p>
8
New cards

What is the LUT?

Series of mathematical equations that are used for postprocessing in radiography

Used as a reference to evaluate raw image data and make corrections to it

<p>Series of mathematical equations that are used for postprocessing in radiography</p><p>Used as a reference to evaluate raw image data and make corrections to it</p>
9
New cards

What is automatic rescaling?

Software automatically adjusting image's grayscale levels based off of histogram analysis in order to produce a normal-appearing radiograph for that view

10
New cards

What is a threshold algorithm?

Computer's method of finding S-min and S-max depending on the number of pixels in each level of gray, defining the region of useful data

11
New cards

What is histogram analysis?

A histogram of the image taken is generated and then compared to an ideal histogram based off of the LUT for that view

Occurs before edge detecting and rescaling

<p>A histogram of the image taken is generated and then compared to an ideal histogram based off of the LUT for that view</p><p>Occurs before edge detecting and rescaling</p>
12
New cards

Characteristic curve

Representation of the response of a screen film radiograph to exposure on a histogram

<p>Representation of the response of a screen film radiograph to exposure on a histogram</p>
13
New cards

What is the difference between characteristic curve when using film versus when using digital?

Film: curve was steeper and narrower, could not recognize exposure that was too low or too high, smaller dynamic range

Digital: a straight line instead of a curve (because digital is much more sensitive and efficient), wider dynamic range

<p>Film: curve was steeper and narrower, could not recognize exposure that was too low or too high, smaller dynamic range</p><p>Digital: a straight line instead of a curve (because digital is much more sensitive and efficient), wider dynamic range</p>
14
New cards

What causes histogram errors?

Anything that presents a great variation from the normal anatomy can throw off histogram analysis (prosthetic, lead, etc.)

May result in large spikes in histogram, which will throw off Smin

Not directly related to technique

15
New cards

What is dark masking?

Puts a dark area behind image

16
New cards

What is image reversal?

Reverses grayscale on image

17
New cards

What is window leveling?

Average brightness of an image

<p>Average brightness of an image</p>
18
New cards

What is another term for window level?

Center (because it refers to the center point of the entire grayscale range)

19
New cards

What is another term for window width?

Sometimes referred to only as window

20
New cards

What is a gradient curve?

Line on a histogram that follows the change of densities on an image. Brightness is represented by length from left to right and contrast is represented by steepness. Curve is adjusted to fit ideal contrast during rescaling.

<p>Line on a histogram that follows the change of densities on an image. Brightness is represented by length from left to right and contrast is represented by steepness. Curve is adjusted to fit ideal contrast during rescaling.</p>
21
New cards

What is gradation?

A gradual passing from one tint or shade to another

22
New cards

What is gradient processing?

When the gradient curve is rescaled to match the gradient curve of the LUT

(rescaled to Qmin and Qmax)

23
New cards

What is smoothing?

Post-processing technique that suppresses visible mottle in an image, reduces noise, and makes edges appear less harsh

Should not be used with severe quantum mottle

<p>Post-processing technique that suppresses visible mottle in an image, reduces noise, and makes edges appear less harsh</p><p>Should not be used with severe quantum mottle</p>
24
New cards

What is the downside to using smoothing?

Can cause loss of contrast and detail

25
New cards

What is edge enhancement?

A post-processing technique that improves visibility of small, high-contrast structures

<p>A post-processing technique that improves visibility of small, high-contrast structures</p>
26
New cards

What is the downside to using edge enhancement?

Image noise is slightly increased

Halo effect

27
New cards

What is the Halo effect?

Black line along structure due to too much edge enhancement

28
New cards

What is spatial location domain?

Assigns anatomical structures to x and y coordinates on an image

29
New cards

The image smoothing feature should not be applied to an image which already has a very low ____________

Spatial resolution

30
New cards

For digital radiographs, spatial resolution should be at least __________.

2.5 LP/mm

31
New cards

What controls slope of gray scale curve?

Gradient curve

32
New cards

Narrow window width results in:

Higher contrast

33
New cards

Which domain uses X and Y coordinates?

Spatial location

34
New cards

An algorithm designed to accentuate soft tissue densities will locate the _______ on the histogram.

VOI

35
New cards

Exposure field recognition is normally done as part of _______________.

Histogram analysis

36
New cards

___________ is best described as an attempt by the computer system to make a low-contrast raw digital image appear like a conventional radiograph prior to any post-processing.

Rescaling

37
New cards

Which portion of the histogram corresponds to the lungs in a chest radiograph?

Left portion of the curve (or whatever side is the darker side)

38
New cards

What is an exposure histogram?

Graph showing gray levels in range of exposure

<p>Graph showing gray levels in range of exposure</p>
39
New cards

What does X and Y represent on an exposure histogram?

X axis (horizontal) represents tones, Y axis (vertical) represents number of pixel in each tone

<p>X axis (horizontal) represents tones, Y axis (vertical) represents number of pixel in each tone</p>
40
New cards

What is exposure field recognition?

The system identifies the part of the image that contains actual anatomy and ignores things like background, shields, etc

41
New cards

What is the region of interest?

Area on histogram that has optimal exposure levels

Between Smin and Smax

Contains values of interest (VOI)

42
New cards

What is S-min?

The lowest useful signal (brightest area of anatomy)

<p>The lowest useful signal (brightest area of anatomy)</p>
43
New cards

What is s-max?

The highest useful signal (darkest area of anatomy)

<p>The highest useful signal (darkest area of anatomy)</p>
44
New cards

What is the S-ave?

Average of S-min and S-max

Where exposure indicator comes from

<p>Average of S-min and S-max</p><p>Where exposure indicator comes from</p>
45
New cards

What are values of interest (VOI)?

Range of densities that a computer algorithm selects in order to accentuate particular anatomy

Within region of interest

<p>Range of densities that a computer algorithm selects in order to accentuate particular anatomy</p><p>Within region of interest</p>
46
New cards

How does a raw digital image appear?

It appears washed out and with low contrast

(Many shades of gray due to digital's linear exposure response)

47
New cards

What is an exposure field recognition?

Process in which computer distinguishes the raw data that is representative of anatomy within the exposure field from that which is outside of the exposure field so that proper automatic rescaling can occur

48
New cards

What are annotations?

Digitally labeling images during postprocessing (adding markers, position/projection, time, etc)

49
New cards

What is image stitching?

Multiple images are attached to one another to form one image

50
New cards

What is window width?

Contrast of an image or length of the grayscale

<p>Contrast of an image or length of the grayscale</p>
51
New cards

What does a steeper gradient curve mean?

The image has higher contrast, shorter gray scale

52
New cards

Still learning (51)

You've started learning these terms. Keep it up!