SN
|
Symbols
|
Explanation
|
1 |
akiec
|
Actinic keratoses and intraepithelial carcinoma |
2 |
bcc
|
Basal cell carcinoma |
3 |
bkl
|
Benign keratosis-like lesions |
4 |
df
|
Dermatofibroma |
5 |
mel
|
Melanoma |
6 |
nv
|
Melanocytic nevi |
7 |
vasc
|
Vascular lesions |
8 |
|
Accuracy |
9 |
R |
Recall |
10 |
P |
Precision |
11 |
|
F1-Score |
12 |
M |
Total models that were used in the system |
13 |
D |
Total datasets that were used in the system |
14 |
m |
Current model that is being studied |
15 |
d |
Current dataset that is being studied |
16 |
|
Denotes the mth model’s prediction score containing the prediction of each class |
17 |
|
Denotes the mth model’s attention output score containing the prediction of each class |
18 |
|
Denotes the mth model’s attention weight |
19 |
C |
Denotes the number of classes in the multiclass framework |
20 |
|
Is the final attention-enabled and ensemble-based model’s output |
21 |
|
Accuracy of model “m” over all D datasets over the K10 protocol |
22 |
|
Accuracy achieved over the dataset “d” over all M Models over the K10 protocol |
23 |
|
Overall system accuracy over M models and D datasets |
24 |
|
AUC of model m summarized over all D datasets |
25 |
|
AUC achieved over dataset d over all M Models |
26 |
|
Overall system AUC over M models and D datasets |
27 |
|
Standard Deviation of the system |
28 |
|
Mean Reliability Index of the system |
29 |
|
Mean Misclassification value |
30 |
|
Mean of Misclassification of ith image over all AI models |
31 |
|
Total number of images for misclassification probability |