1. Intro to Imaging & Digital Image Processing

Defining Imaging

• Modality: The difference between different bioimaging methods and machines (ex: CT and MRI are different )
  • Four necessary components of a modality: source (illumination), camera (detector), digitizer (frame grabber), imaging processing unit
  • Imaging processing unit (hardware and/or software)
  • acquisition (takes in the data and understands it)
  • preprocessing (combines information from multiple points)
  • segmentation (identifying components, facial recognition)
  • & more
• An image is a 2D representation of a physical quantity as rendered by an imaging modality
  • X-ray attenuation (projection x-ray yor CT)
  • Proton density (MRI)
  • Acoustic reflectivity (ultrasound)
  • An image represents a “finite-thickness” plane within a volume of interest
• Types of imaging
  • Anatomical: Imaging that represents structure/composition of objects (e.g. CT imaging)
  • Functional: Imaging that represents function/physiology of an organ (e.g. PET scans)
  • Projection: Imaging that shoes a single planar representation (e.g. x-ray)
  • Tomographic: Imaging that shows cross-sectional representation (e.g. CT imagings)
• Imaging mechanisms
  • Transmission: The imaging mechanism by which information comes from what travels through the body (e.g. x-ray)
  • Reflection: Transmission: The imaging mechanism by which information that comes from what reflects back from the body (e.g. ultrasound)

Modality Comparison

• Ionizing is when the energy we work with is higher than others, and electrons in the atoms can bump up to unsafe levels; this is something we want to avoid (can lead to cancer)

  

Digital Imaging

• Digital images are digital files saved on a computer
• 2D arrays of “picture elements” called pixels
  • Voxels are for 3D elements
• Image size = width x height
• Real-world image size (or FOV) is (Ncolumn x pixel width) x (Nrow x pixel height)
• Resolution: Number of pixels per square inch
• Image Pixels
  • Addressed with x,y coordinates
  • Top left corner is (1, 1)
  • (coumn, row)
• Storage type
  • Pixel values depend on the storage type
  • Grayscale images are NOT called black and white
  • 8-Bit: Greyscale images with values from 0 to 2^8 minus 1
  • 16-Bit: Greyscale images with values from 0 to 2^16 minus 1
• Color images: each pixel can have 4 values
  • 1 value per pixel – e.g. indexed image
  • 3 values per pixel e.g. 3x1 bytes – RGB, 3x2 bytes – RGB, …
  • 4 values per pixel RGB e.g. 4x1 byte – RGB_Alpha, …

Image Processing

• Enhancement/restoration of image info for human reading
• Segmentation
• Characterization
• Representation of images for machine analysis
• Visualization
• Processing of image data for storage
• Processing of image data for transmission
• Matlab
  • Load the image and info
  • imread()
  • iminfo()
  • Display image
  • imtool()
  • imshow()
  • imagesec()
  • image()
  • imshowpair()
  • Perform needed operation
  • Display and evaluate results
  • Save resulting image
  • imwrite()