Geometric Model Reconstruction - Computational Hemodynamics: Theory, Modelling and Applications - page 65

3.6 Review Questions

1. Why are the scanned images from CT referred to as 'slices' ?
2. What is the significance of water content in the body for MRI scans?
3. For the following anatomy, which scan (MRI or CT) would you use: soft tissue,
bone structure, mucous walls.
4. For the respiratory airway, what are the main issues to consider when using
MRI and CT scans?
5. What is the Hounsfield scale?
6. What is the difference between a pixel and a voxel?
7. What does manual segmentation involve?
8. What is the problem with using a fully automated segmentation?
9. What is the threshold greyscale range? What are you looking to segment if you
had a range of 2000-3072 HU?
10. What is the main property of pixels in an image where an edge occurs?
11. Why do images that use the LoG and Canny edge detection schemes need to be
smoothed?
12. What does the term ' zero crossing ' mean?
13. The Roberts edge detection algorithm uses the following mask:

What would the approximate first derivative of a pixels intensity be?
14. Give a brief description of a region-based segmentation algorithm. How does it
differ from edge detection algorithms?
15. What is a 'dirty' CAD geometry?
16. Why is it problematic in accessing CAD models between different software?