1 / 0

Medical Imaging

Medical Imaging. Mohammad Dawood Department of Computer Science University of Münster Germany. What is medical imaging? Medical imaging is the process of acquiring images without or with minimal invasion for the purpose of detecting, diagnosing, quantifying or treating a disease.

dacia
Télécharger la présentation

Medical Imaging

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Medical Imaging Mohammad Dawood Department of Computer Science University of Münster Germany
  2. What is medical imaging? Medical imaging is the process of acquiring images without or with minimal invasion for the purpose of detecting, diagnosing, quantifying or treating a disease. Techniques and methods from image processing are used to assist the clinicians.
  3. Structure of the Course Basics of Image processing Medical Image modalities Reconstruction Registration Segmentation Enhancement
  4. Image processing Signal processing with an image as an input and an image or a set of features as output. Definitions Image Domain In the discrete case
  5. Classical methods of image processing include Grayscale transformations Color spaces Filtering Edge detection Morphological operations
  6. Grayscale transformations The human eye can distinguish between different colors with estimates ranging from 100,000 to 10 million!
  7. Michelson contrast : Weber contrast:
  8. Grayscale Transforms
  9. Grayscale transformations Three of the most common grayscale transforms are: Linear Logarithmic Power law Point operations
  10. Linear color domain transform X-Ray Mammogram
  11. Power law MRI of Spinal cord
  12. Power law CT of Head
  13. Histogram Histogram function : Probability function: Cumulative histogram:
  14. Histogram Equalization MRI of Spinal cord
  15. Histogram equalization Mammograms
  16. Adaptive/Local Histogram Equalization
  17. Local Histogram Equalization
  18. Use of color spaces
  19. Use of different color spaces The continuous spectrum visible to human eyes
  20. Use of different color spaces RGB (Red, Green, Blue)
  21. Use of different color spaces RGB (Red Green Blue) Cardiac PET
  22. Use of different color spaces HSV (Hue, Saturation, Value)
  23. Use of different color spaces HSV (Hue, Saturation, Value) S=1, V=1 V=1 S=1 Cardiac PET
  24. Using different spectrums Cardiac PET
  25. Fourier Transform Euler’s formula: Fourier transform: Inverse Fourier transform:
  26. Fourier Transform Respiratory signal
  27. Fourier Transform Convolution theorm
  28. Spatial filtering
  29. Spatial connectivity 2D - 4 connectivity - 8 connectivity 3D - 6 connectivity - 18 connectivity - 26 connectivity
  30. Spatial filtering (local operators) Filters are used in image processing for various purposes e.g. noise reduction, edge detection, pattern recognition. * 1/9 Applied only to red cell f h f* (0*1+7*1+3*1-1*1+8*1+3*1+4*1+0*1+3)*1/9 = 3
  31. Noise reduction Averaging filter * *1/9 = Applied only to red cells Cardiac PET, averaging with 5x5
  32. Median filter Median = Middle value of the set Example - given S = {1, 5, 2, 0, -3, 8, 0} - sort S = {-3, 0, 0, 1, 2, 5, 8} median(S)= 1 What happens if |s| is even? - given S = {1, 5, 2, 0, -3, 8, 0, -5} - sort S = {-3, -5, 0, 0,1, 2, 5, 8} median(S)= 0.5
  33. Noise reduction Median filter * median filter = Applied only to red cells
  34. Noise reduction Gaussian filter Gauss function is defined as:
  35. Noise reduction Comparison Original Averaging (5x5) Median(5x5) Gaussian (5x5)
More Related