Digital Age Ch.28

Chapter 28 - Creating the Digital Image

Objectives

• Upon completion of this chapter, you should be able to:
  1. Describe the aspects of a digital image matrix and how it impacts spatial resolution.
  2. Relate pixel size to the displayed field of view and matrix size.
  3. Define the three steps in digitizing any analog image.
  4. Explain the relationships between bit depth, dynamic range and image gray scale in providing contrast resolution.
  5. Describe the nature of voxels for CT, CR and DR imaging and how the x-ray attenuation coefficient for each is translated into the gray levels of pixels.
  6. Describe the development and limitations of contrast resolution and spatial resolution for digitized radiographic images.
  7. Explain how the selection of specific procedural algorithms impacts the displayed image.
  8. Fully define window level and window width and how they translate into displayed image brightness and gray scale.
  9. Describe the components and function of the PACS, RIS and HIS, and the DICOM standard.
  10. Define the types, characteristics, and proper use of workstations and display stations.

The Nature of Digital Images

• All digital images, whether photographic, radiographic, or fluoroscopic, consist of a matrix of numeric values stored in computer memory.
• The matrix is a pattern of cells arranged in rows and columns.
• Each cell corresponds to a specific location within the image and can be identified by its column and row designations.
• In radiographic images, the numerical value stored for each cell represents brightness (density) assigned to that location.
• This brightness level ranges from "pitch black" (value 0) to "blank white".

Example of Image Representation

• Matrix Representation:
  • Bone tissue in the femur (very light gray) = numeric value of 555.
  • Soft tissue in the thigh (dark gray) = numeric value of 11.
  • Background (pitch black) = pixel value of 0.
  • This arrangement illustrates how denser tissues are represented by higher numbers, building an image of bone within thigh tissue.

Visualization of Matrices

• Digital Image Examples (see Figures 28-2 and 28-3):
  • Two matrices with higher values around the center display distinct anatomical parts not immediately apparent without close inspection.

Pixel and Matrix Size

• Each cell within a digital image is called a pixel (short for "picture-element").
• Example of matrices:
  • Matrix A: 6 pixels height x 6 pixels width = 36 total pixels.
  • Matrix B: 12 pixels height x 12 pixels width = 144 total pixels.
• Larger matrices contain smaller pixels, improving spatial resolution.
• Spatial resolution can be measured in terms of spatial frequency with units of line-pairs per millimeter (LP/mm).
  • At least two pixels are required to record a line pair with differing densities.

Spatial Resolution Limits

• Higher pixel density (smaller pixel size) yields sharper resolution.
• Historically, spatial resolution improved significantly from the 1990s onward:
  • High-speed film/screen systems achieved 8-10 LP/mm.
  • Modern digital systems approach these values but generally remain inferior in spatial resolution compared to older film systems (10-12 LP/mm).

Mathematical Relationship of Spatial Frequency

• Formula: $SF = \frac{2}{PS}$
  • SF = spatial frequency (LP/mm), PS = pixel size (mm).

Practice Exercise

• For a pixel size of 0.3 mm, calculate the associated resolution in LP/mm:
  • Solution:
  • $SF = 2(0.3) = 1.6 ext{ LP/mm}$
  • For a pixel size of 0.2 mm, $SF = 2(0.2) = 2.5 ext{ LP/mm}$.

Digitizing an Analog Image

• Three steps involved:
  1. Scanning: The image is divided into a matrix of small cells (pixels).
  2. Sampling: Intensity measurements of light or radiation from each designated pixel area.
  3. Quantization: Assigning a value (discrete gray level) to each pixel from a predetermined dynamic range.

Definitions and Concepts

• Bit Depth: The exponent of base 2 representing the number of gray levels for each pixel:
  • 6 bits = 64 levels; 8 bits = 256 levels.
• The dynamic range is the range of gray levels available for constructing images, influenced by the bit depth.
• Maximum Dynamic Range: Set by the system software processing the image data and affects overall image quality, including gray scale.

X-Ray Attenuation in Digital Images

• All imaging modalities rely on measuring x-ray attenuation as the beam passes through the patient.
• Attenuation coefficient determines the level of brightness assigned to a pixel.
• Voxels (Volume Elements) used in CT scans reflect three-dimensional tissue volumes corresponding to pixel values.
  • In CT, averaging the attenuation coefficients of tissues determines pixel values.

Enhancement of Contrast Resolution

• Digital imaging allows manipulation of pixel gray scale values post-acquisition.
• This contrasts with analog radiography, which required a minimum of 10% subject contrast within adjacent structures to image effectively.
  • Digital systems can resolve differences as low as 1% due to enhanced contrast resolution capabilities.

Procedural Algorithms and Their Applications

• Procedural algorithms are pre-set dynamics controlling image output for specific anatomical areas, adjusting brightness and gray scale to optimize detail visualization.

Windowing - Control Functions

• Window Level (WL): Adjusts overall brightness; sets the average gray level of the image.
• Window Width (WW): Controls the range of gray scale in the displayed image; increasing width increases visible shades but lowers contrast.

Practical Considerations in Image Viewing

• Workstations allow full image manipulation and storage, while display stations are limited to viewing.
• Effective resolution, ambient lighting conditions, and proper monitor capabilities significantly enhance image diagnosis.

Summary of Key Concepts

1. Larger digital image matrix = smaller pixel size = higher spatial resolution.
2. Pixel size is related to field of view and matrix size.
3. Analog data is continuous; digital data is discrete.
4. The digitization process involves scanning, sampling, and quantizing.
5. The dynamic range and bit depth influence image gray scale and resolution.
6. Subject contrast resolution is significantly improved in digital imaging compared to analog methods.
7. Image processing protocols are critical for optimizing radiographic procedures.
8. Workstations differ from display stations in functionality and resolution capabilities.