chapter 2 -- digital image characteristics

digital imaging

imaging acquisition process that produce an electronic image that can be viewed and manipulated on a computer

digital image

a numerical representation of an image via a set of picture elements known as pixels

• can be processed and manipulated with softwares

• can be stored electronically (soft copies)

• can be dictated, reviewed, and transmitted virtually

characteristics of a digital image

• a digital image begins as an analog signal

• through computer data processing, the image becomes digitized and is sampled multiple times

  • convert part of analog into digital signals

    • computer can only read digital data

• ADC

analog signal/data

• a continuously changing signal

• converted into digital data

digital signal/data

• recorded as multiple numeric values and are divided into an array of small elements that can be processed in many different ways

• begins as an analog signal; represents a sequence of discrete values

ADC

analog-to-digital converter

• through computer data processing, the image becomes digitized and is sampled multiple times

DAC

digital to analog converter

image domains

• digital images are represented into 2 domains:

  • spatial location domain

  • spatial frequency (or frequency) domain

  • the fourier transformation(FT): or reverse FT (just different directions)

spatial location domain

• based on a matrix that has specific locations for each pixel

• what radiographers and radiologists see

spatial frequency (or frequency) domain

• based on the number of cyclers per unit length (frequency)

• what physicists and engineers see to make sure machine is working properly

the fourier transformation(FT): or reverse FT (just different directions)

the mathematical algorithm that is applied to change an image from the spatial location domain to the spatial frequency domain

small object/high contrast

high frequency

large object/low contrast

low frequency

pixel

• picture element; the smallest element in a digital image

  • no contrast in each pixel (uniform)

• magnification of a digital picture reveals small squares

• features determine spatial resolution and contrast resolution

spatial resolution

detail; sharpness; recorded detail — the degree of geometric sharpness or accuracy of structural lines actually recorded in the image

• the ability of the imaging system to demonstrate small details of an object

• for better resolution, you want smaller pixels

contrast resolution

the ability of imaging systems to distinguish and display a range of attenuation characteristics from different tissues within the body

• how much information is held in pixel

• higher the range, more grayscales are displayed, better resolution

• refers the the ability of the digital system to display subtle changes in the shades of gray

pixel size

• directly affects image spatial resolution

  • the smaller the pixel, the higher the spatial resolution

• is related to the size of matric and size of field of view

matrix

• a square arrangement of numbers in columns and rows

• in digital imaging, the number corresponds to discrete pixel values

• each box within corresponds to:

  • a specific location in the image

  • a specific area of the patient’s tissue

• the image is digitalized both by position (spatial location) and by intensity (gray level)

• the typical number of pixels in one ranges from about 512 × 512 to 1024 × 1024 and can be as large as 2500 × 2500

• the size determines the size of the pixels

• the larger, the larger the image file size

  • important to know for archival needs

matrix size

for a given field of view, the larger the matrix size, the greater the number of smaller individual pixels, increasing the number of smaller pixels will improve the quality of the image

field of view (FOV)

• synonymous with x-ray field

  • it is the amount of body part or patient included in the image

  • the lager, the more area is imaged

• changes in the FOV does not affect the size of the matrix

  • changes in matrix affect pixel size

  • as the matrix increases and the FOV remains the same, the pixel size must decrease to fit into the matrix

pixel size, matrix size, and FOV

• a relationship may exist between the size of the pixel, the size of the matrix, and the field of view

  • matrix size can be changed without affecting the FOV

  • the FOV can be changed without affected the matrix size

  • a change in either the matrix size of the FOV will change the size of the pixels

pixel size =

FOV / matrix size

binary code

• machine language of 1s and 0s

• processing is performed by a series of transistors, which are switches that are on or off

• the computer’s transistors can be switched on and off millions of times in a second

• two symbol alphabet

  • off - 0

  • on - 1

• numbers function on base 2

  • example (10-based to 2-based)

• how much information is stored in each unit location is determined by 2 base system

  • red: 11111111 00000000 00000000

  • green: 00000000 11111111 00000000
    blue: 00000000 00000000 11111111

on

if the circuit is closed and current passes through, it is assigned a value of 1

off

if the circuit is open and current does not pass through, it is assigned a value of 0

bit

binary digital; a single unit of data

• a 1 or 0

byte

• made up of 8 bits

• the amount of memory needed to store on alphanumeric character

  • 8 bit word — “A” = 01000001

• memory is defined by the number of bytes (1024 conversion factor)

  • kilobyte, megabyte, gigabyte, terabyte

1 =

1

2 =

10

3 =

11

4 =

100

0 =

0

bit depth

• the number of bits or pieces of information within each pixel

  • determines how much information can be stored

• 8 bits = 1 byte

• also determines how many gray tones can be represented in each pixel

  • calculated by 2 to the power of the bit depth

• bit depth range for digital systems is 10-16 or 2¹⁰, 2¹², or 2¹⁶ shades of gray (1024 - 65,536)

• each pixel can have a gray level between 1 (2⁰) and 65,536 (2¹⁶)

exposure indicators (EI)

• numeric value that provides feedback on the amount of radiation that reached the digital image receptor (not the patient)

• a quality control tool used to:

  • verify appropriate exposure

  • reduce repeat imaging

  • optimize radiation dose

deviation index

• abbreviated as DI

• simply, the difference between the actual exposure (K_IND) and the target exposure (K_TGT)

  • expressed in a logarithmic fashion

    • DI = 10 x log10[K_IND/K_TGT(b,v)]

• technologists must evaluate images, and if they are unsure of its quality, they should consult with a radiologist

  • radiologists have a higher quality display than what is normally used for technologist workstations

• a perfect image according to the DI has a DI value of 0.0

• this ± 1 changes the K_IND by +25% or -20%

  • +1 is ~125% of the intended exposure

  • -1 is 80% of the intended exposure

• several variances could cause the reading of the pixel values (Q) to be off:

  • a protheses within the image

  • gonadal shielding within the image

  • failure of the system to recognize the collimated border

  • an unexpected body part in the image

• any of the above can cause the DI to fluctuate

  • technologists should consult with the radiologist regarding the above image variances

• technologists should continue to use image noise as the true determining factor of image acceptance

  • these standard values should be used as a guide

  • signal to noise ration

air kerma

• kerma is kinetic energy released per unit mass [of air]

• the measurement of radiation energy (joules or J) absorbed in a unit of air (kg)

• most often expressed as J/kg or gray (Gy)

indicated equivalent air kerma

• abbreviated as K_IND

• the measurement of the radiation that was incident on the IR for that particular exposure

• simply stated, K_IND is the amount of exposure on the IR

target equivalent air kerma value

• abbreviated as K_TGT

• a set of specific values that represents an optimal exposure for each body part and view

  • values are set by the manufacturer or system user

• these values are listed in a table in the system by body part (b) and view (v), K_TGT (b, v)

if the DI is negative

the image is underexposed

• to raise the DI +1, increase technique by 25%

if the DI is positive

the image has been overexposed

• to decrease DI -1, decrease technique by 20%

image quality characteristics

brightness, contrast resolution, spatial resolution

brightness

• refers to the appearance of the image on a display monitor

• no technical adjustment during image acquisition will change the brightness of the image; it is strictly a display quality

modulation transfer function (MTF)

• the ability of a system to record available spatial frequencies

  • quantifies the contribution of each system component to the overall efficiency of the entire system (overall resolution)

• measures the accuracy of an image compared to the original object on a scale of 0-1 or 0 to 100%

  • measures the trueness: fidelity

• the sum of the overall contribution from every component can not be greater than the system as a whole

  • it is a ratio of the image to the object so that a perfect system would have an MTF of 1 or 100%

  • if anything interferes with the performance of the whole system, the MTF decreases. Such as light spreading, artificial hip, pathology, Unfiltered or corrected noise, etc.

• in digital detectors when an x-ray photon energy excites a phosphor/scintillation layer:

  • the phosphor produces light

  • spreading out of the light will always occur

  • light spread reduces system efficiency

• the more light spread, the lower the MTF

• in this imaging system, it images large structure more accurately (truthgully)

• however, it is difficult to image small structures

noise

• anything that interferes with the formation of the image

• radiographic noise occurs during the acquisition of the image

  • equipment noise is caused by detector elements or nonuniform detector responses

  • quantum noise — results in quantum mottle

• signal-to-noise ratio

• contrast-to-noise ration

signal-to-noise ratio (SNR)

• a measurement of the image signal in a given region to the background noise

• as SNR increases, noise decreases, but exposure increases to the patient

• how much noise can be tolerated by the radiologist

contrast-to-noise (CNR)

a measurement of the contrast between tissue of interest and the background noise

dynamic range

the range of exposure IR is able to detect

exposure latitude

the exposure range produces the diagnostic image

• linear in digital system; S in film system

• wider dynamic range, also wider exposure latitude in digital imaging system

• depends on the imaging detector

  • the higher the dynamic range of the detector, the more values can be detected

detective quantum efficiency (DQE)

• describes how efficiently a system converts the x-ray input signal into a useful output image

• is a measurement of the percentage of x-rays that is absorbed when they hit the detector

• systems with higher quantum efficiency can produce higher quality images at a lower dose

• amorphous selenium, amorphous silicon TFT, charge-coupled device (CCD), and complementary metal oxide semiconductor detector technology have increased DQE over PSP

• amorphous selenium detectors have the highest DQE because they do not have the light conversion step

  • no light spread, no light to blur the recorded signal output, higher quality images are produced

  • disadvantage: such as more expensive than others

• comparing others:

  • the higher fill factor, the more radiation can be detected, and the greater the signal generated, the higher the DQE