Table 1.
Portrayal of the dataset.
Fig 1.
Sample Images of Different Classes (a) Actinic Keratosis (b) Basal Cell Carcinoma (c) Benign Keratosis (d) Dermatofibroma (e) Melanoma (f) Nevus (g) Vascular Lesions.
Fig 2.
Instance distribution for each class.
Fig 3.
Schematic representation of methodology.
Fig 4.
Sample images of the images after augmentation.
(a) Original Sample, (b) Rotated Sample, (c) Width_shifted, (d) Height_shifted, (e) Zoomed, (f), Horizontal_flipped, (g) Vertical_flipped.
Fig 5.
CTL architecture.
Fig 6.
Feature extraction after activation of each layer (One image as example).
Fig 7.
IGPA in Level—n.
Fig 8.
Organization of ML-IGPA.
Table 2.
Trainable parameters for different algorithms.
Table 3.
Performance metrics of DenseNet121 approaches.
Table 4.
Performance metrics of DenseNet169 approaches.
Table 5.
Performance metrics of DenseNet201 approaches.
Table 6.
Performance metrics of MobileNet approaches.
Table 7.
Performance metrics of MobileNetV2 approaches.
Table 8.
Performance metrics of MobileNetV3Large approaches.
Table 9.
Performance metrics of InceptionV3 approaches.
Table 10.
Performance metrics of InceptionResnetV2 approaches.
Table 11.
Performance metrics of Xception approaches.
Table 12.
Performance metrics of DN approaches.
Table 13.
Performance metrics of MN approaches.
Table 14.
Performance metrics of IX approaches.
Table 15.
Performance metrics of DMIX approaches.
Table 16.
Performance metrics of Single-Levelled IGPA.
Fig 9.
Confusion matrix obtained by SL-IGPA.
Fig 10.
Confusion matrix obtained by ML-IGPA (All classifiers).
Fig 11.
Confusion matrix obtained by ML-IGPA (Best 3 classifiers).
Fig 12.
ROC-AUC curve obtained by SL-IGPA.
Fig 13.
ROC-AUC curve obtained by ML-IGPA (All classifiers).
Fig 14.
ROC-AUC curve obtained by ML-IGPA (Best 3 classifiers).
Fig 15.
GradCAM generation by the model for each class.
(a) GradCAM for AK, (b) GradCAM for BCC, (c) GradCAM for BKL, (d) GradCAM for DF, (e) GradCAM for MEL, (f) GradCAM for NV, (g) GradCAM for VASC.
Fig 16.
GradCAM visualization for architecture explainability (Example by DN121).
(a) Original, (b) CACNN, (c) SEACNN, (d) SACNN.
Table 17.
Performance metrics of IGPA without TA.
Table 18.
Performance metrics of Softmax Averaging of all classifiers.
Table 19.
Performance metrics of Softmax Averaging of best 3 classifiers.
Table 20.
Performance metrics of Single Level Softmax Averaging.
Table 21.
Performance metrics of Majority Voting of all classifiers.
Table 22.
Performance metrics of Majority Voting of best 3 classifiers.
Table 23.
Performance metrics of Single Level Majority Voting.
Table 24.
Performance metrics of Weighted Averaging of all classifiers.
Table 25.
Performance metrics of Weighted Averaging of best 3 classifiers.
Table 26.
Performance metrics of Single Level Weighted Averaging.
Table 27.
Comparison of our proposed model with other existing models.