Medical Image Description Based on Multimodal Auxiliary Signals and Transformer

Table 3

Performance of different feature extraction networks.

	BLEU1	BLEU2	BLEU3	BLEU4	METEOR	L
ResNet-101	0.505	0.318	0.219	0.159	0.195	0.383
ResNet-152	0.489	0.310	0.219	0.157	0.210	0.375
ResNet_101_32 × 8d	0.493	0.306	0.203	0.137	0.198	0.366
wide_ResNet-101_2	0.499	0.309	0.206	0.143	0.198	0.346

The bold values indicate that the model performance of the algorithm is optimal in a certain type of dataset.