Classification algorithm of retinopathy based on attention mechanism and multi feature fusion

Table 1. Performance of ablation experiments in the IDRID dataset
View table
View in Article
Table 1. Performance of ablation experiments in the IDRID dataset
Model QWK/% Acc/% Se/% Sp/%
M1 84.83 71.84 88.48 88.23
M2 85.08 74.75 89.80 94.11
M3 87.70 77.66 92.75 89.80
M4 89.59 76.69 96.10 88.23
AMMF 90.38 78.64 95.65 91.17

Table 2. Performance of different models in the IDRID dataset

View table

View in Article

Table 2. Performance of different models in the IDRID dataset

Methods	Backbone	QWK/%	Acc/%	Se/%	Sp/%
Bhardwaj et al.^[18]	ResNet-50	－	79.46	82.85	76.98
Wu et al.^[19]	ResNet-18	－	56.19	64.21	87.39
Song et al.^[20]	ResNet-50	86.54	76.69	89.85	88.00
Shi et al.^[21]	Efficientnet-b5	87.63	79.06	－	－
Ours	ResNeSt-50	90.38	78.64	95.65	91.17

Table 3. Performance of different models in the Eye-PACS dataset

View table

View in Article

Table 3. Performance of different models in the Eye-PACS dataset

Methods	Backbone	Acc/%	AUC/%	Se/%	Sp/%
Wu et al.^[19]	ResNet-18	83.10	89.00	53.99	91.22
Song et al.^[20]	ResNet-50	82.25	87.08	61.27	90.21
Kamothi et al.^[22]	Modified VGG-16	77.60	－	77.00	－
Thota et al.^[23]	VGGNet	74.00	80.00	80.00	65.00
Saxena et al.^[24]	InceptionResNetV2	76.68	79.51	50.77	91.25
Ours	ResNeSt-50	84.41	90.36	77.85	97.05

Tools

Get Citation

Copy Citation Text

Liming Liang, Xin Dong, Renjie Li, Anjun He. Classification algorithm of retinopathy based on attention mechanism and multi feature fusion[J]. Opto-Electronic Engineering, 2023, 50(1): 220199

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites