Novel Discrete Compactness-Based Training for Vector Quantization Networks: Enhancing Automatic Brain Tissue Classification

Table 13

Associations between winner classes and tissue characterizations for network topology 𝜏₂. There is a set of If-Then rules for each evaluation set.

First set		Second set
If winner class is	Then characterize mask as	If winner class is	Then characterize mask as
0	Background-0	0	Background-0
1	Background-0	1	Skin-6
2	CSF-1	2	Fat-4
3	Background-0	3	Skull-7
4	CSF-1	4	Fat-4
5	Background-0	5	Skull-7
6	Background-0	6	Skin-6
7	Background-0	7	Background-0
8	Background-0	8	Background-0
9	Skull-7	9	Skin-6
10	Background-0	10	Background-0
11	Background-0	11	Background-0
12	Background-0	12	Skin-6
13	Background-0	13	Skin-6
14	Background-0	14	Background-0
15	CSF-1	15	Fat-4
16	Background-0	16	Skull-7
17	Grey Matter-2	17	Grey matter-2
18	Background-0	18	Background-0
19	Background-0	19	Skin-6