Table 3.
Accuracy results for classification of anatomical types for IEEE ISBI 2015 Challenge Dataset 1.7
| Reference | IEEE ISBI 2015 Challenge Dataset 1 | ||||||
|---|---|---|---|---|---|---|---|
| Confusion matrix (proposed) | Diagonal average (proposed, %) | Diagonal average28,33–35 (%) | Diagonal average37 (%) | ||||
| ANB |
Type 1 (%) | Type 2 (%) | Type 3 (%) | 61.47 |
59.42 |
64.99 |
|
| Type 1 | 46.15 | 10.26 | 43.59 | ||||
| Type 2 | 40.00 | 51.43 | 8.57 | ||||
| Type 3 |
11.84 |
1.32 |
86.84 |
||||
| SNB |
Type 1 | Type 2 | Type 3 | 70.11 |
71.09 |
84.52 |
|
| Type 1 | 71.43 | 14.29 | 14.29 | ||||
| Type 2 | 33.33 | 58.33 | 8.33 | ||||
| Type 3 |
17.47 |
1.94 |
80.58 |
||||
| SNA |
Type 1 | Type 2 | Type 3 | 63.57 |
59.00 |
68.45 |
|
| Type 1 | 60.00 | 10.00 | 30.00 | ||||
| Type 2 | 34.94 | 56.63 | 8.43 | ||||
| Type 3 |
18.52 |
7.41 |
74.07 |
||||
| ODI |
Type 1 | Type 2 | Type 3 | 75.04 |
78.04 |
84.64 |
|
| Type 1 | 70.97 | 17.74 | 11.29 | ||||
| Type 2 | 26.67 | 73.33 | 0.00 | ||||
| Type 3 |
19.18 |
0.00 |
80.82 |
||||
| APDI |
Type 1 | Type 2 | Type 3 | 82.38 |
80.16 |
82.14 |
|
| Type 1 | 72.5 | 15.00 | 12.50 | ||||
| Type 2 | 15.79 | 81.58 | 2.63 | ||||
| Type 3 |
6.94 |
0.00 |
93.06 |
||||
| FHI |
Type 1 | Type 2 | Type 3 | 65.92 |
58.97 |
67.92 |
|
| Type 1 | 87.88 | 6.06 | 6.06 | ||||
| Type 2 | 66.67 | 33.33 | 0.00 | ||||
| Type 3 |
23.46 |
0.00 |
76.54 |
||||
| FMA |
Type 1 | Type 2 | Type 3 | 73.90 |
77.03 |
75.54 |
|
| Type 1 | 61.76 | 26.47 | 11.76 | ||||
| Type 2 | 10.75 | 86.02 | 3.23 | ||||
| Type 3 |
21.74 |
4.35 |
73.91 |
||||
| MW | Type 1 | Type 2 | Type 3 | 81.31 | 83.94 | 82.19 | |
| Type 1 | 77.78 | 11.11 | 11.11 | ||||
| Type 2 | 11.29 | 87.10 | 1.61 | ||||
| Type 3 | 9.30 | 11.63 | 79.07 | ||||