1. What is the structure of adaptive fuzzy leader clustering (AFLC)?

2. Does AFLC have to initialize like k-means? If not, why?

3. How does AFLC dynamically adjust the number of clusters?

4. What is the difference between deterministic annealing (DA) and simulated annealing (SA)?

5. What is the DA cost function and what does it minimize?

6. What effect does the temperature reduction rate parameter have on DA clustering?

7. How does DA adjust the number of clusters?

8. What does mass-constrained DA mean?

9. What makes MS segmentation different from normal brain segmentation?

10. Judging from the examples given in the chapter, what are the performance differences among AFLC, DA, FCM, and k-means?

11. What is the limitation of clustering segmentation based on image intensity?

12. How is clustering in retinal optic disk/cup and blood vessel segmentation better than regular edge detection techniques?

13. Why is registration necessary in 3-D retinal disk/cup segmentation and how is it done?

14. How is the 3-D optic disk/cup map created?

0 0

Post a comment