Table 22.

Comparative study in terms of the accuracy with the state-of-the-art models

Reference	Model	Accuracy %
Gumaei et al. [10]	Regularized extreme learning machine	94.23
Sajjad et al. [11]	VGG19 with extensive data augmentation	94.58
Anarki et al. [15]	CNN with genetic algorithm	94.20
Swati et al. [13]	VGG19 with fine tuning	94.82
Deepak et al. [16]	GoogleNet with transfer learning	97.10
Alshayeji et al. [17]	Aggregation of two paths from CNN	97.37
Kakarla et al. [18]	Average pooling convolutional neural network	97.42
Kumar et al. [19]	ResNet-50 with Global Average Pooling at the output layer	97.48
The proposed BTC-fCNN model (case 3)	The proposed model with retraining the model in each fold during five folds	98.86