Skip to main content
. Author manuscript; available in PMC: 2013 Oct 1.
Published in final edited form as: Med Image Anal. 2012 May 29;16(7):1415–1422. doi: 10.1016/j.media.2012.04.010

Figure 5.

Figure 5

Browsing result for querying the top K = 5 most-characteristic images per pit pattern. The subset of all images that remains after patient pruning is highlighted. The top row shows a schematic textbook visualization of the Kudo criteria for pit pattern discrimination (cf. [Kudo et al., 1994]): Type I is characterized by normal, round pits, type II by asteroid, stellarorpapillary pits, type III-L by tubular or round pits (usually larger than type I), type III-S by tubular or round pits (usually smaller than type I), type IV by dendritic or gyrus-like pitsandtype V by irregular arrangements, or a compete loss of structure.