Table 2.

Comparative classification performance of our proposed network with the other state-of-the-art methods used in endoscopy.

Authors	Deep network	Accuracy, %	F1, %	mAPa, %	mARb, %
Zhang et al (2017) [44]	SqueezeNet [57]	77.84	76.74	76.77	76.73
Hicks et al (2018) [45]	VGG19 [54]	85.15	85.29	85.88	84.72
Fan et al (2018) [36]	AlexNet [16]	80.08	80.49	80.70	80.28
Takiyama et al (2018) [8]	GoogLeNet [58]	84.59	85.14	85.29	84.99
Byrne et al (2019) [5]	InceptionV3 [55]	87.92	88.45	87.87	89.05
Jani et al (2019) [46]	MobileNetV2 [59]	88.53	88.51	88.34	88.69
Lee et al (2019) [39]	ResNet50 [56]	89.55	90.60	90.70	90.50
Vezakis et al (2019) [40]	ResNet18 [56]	89.95	90.35	90.72	89.99
Owais et al (2019) [10]	CNNc + LSTMd [28,56]	92.57	93.41	94.58	92.28
Cho et al (2019) [42]	InceptionResNet [60]	84.78	84.53	84.15	84.92
Dif et al (2020) [38]	ShuffleNet [61]	89.63	89.14	88.67	89.63
Song et al (2020) [43]	DenseNet201 [27]	92.12	92.42	92.91	91.93
Guimarães et al (2020) [37]	VGG16 [54]	85.72	85.80	86.24	85.37
Hussein et al (2020) [41]	ResNet101 [56]	90.24	91.14	91.52	90.78
Klang et al (2020) [47]	Xception [62]	86.05	84.88	84.19	85.58
Proposed method	DenseNet + LSTM + PCAe + KNNf	96.19	96.99	98.18	95.86

^amAP: mean average precision.

^bmAR: mean average recall.

^cCNN: convolutional neural network.

^dLSTM: long short-term memory.

^ePCA: principal component analysis.

^fKNN: k-nearest neighbor.