Abstract | ||
---|---|---|
Age-Related Macular Degeneration (AMD) and Diabetes Macular Edema (DME) are eye diseases with the highest blinding rate. Optical Coherence Tomography (OCT) is widely used to diagnose different eye diseases. However, the lack of automatic image analysis tools to support disease diagnosis remains a problem. At present, the high-dimensional analysis of OCT medical images using Convolutional Neural Networks (CNN) has been widely used in the fields of visual field assessment of glaucoma and diabetes retinopathy. The method we proposed involves the transfer learning of Inception V3. The experiment includes two stages: (1) Firstly, using SinGAN to generate high-quality image samples and enhance the data; (2) Fine-tune and validate the Xception model generated using transfer learning. The research shows that the Xception model achieves 98.8% classification accuracy on the OCT2017 data set under the condition that the Xception model has the same parameter quantity as the Inception model, to realize a more accurate classification of OCT images of blinding diseases. |
Year | DOI | Venue |
---|---|---|
2022 | 10.1007/978-3-031-13841-6_17 | INTELLIGENT ROBOTICS AND APPLICATIONS (ICIRA 2022), PT IV |
Keywords | DocType | Volume |
OCT, Deep learning, Xception, SinGAN, Image classification | Conference | 13458 |
ISSN | Citations | PageRank |
0302-9743 | 0 | 0.34 |
References | Authors | |
0 | 6 |
Name | Order | Citations | PageRank |
---|---|---|---|
Rong Wang | 1 | 0 | 0.34 |
Yaqi Wang | 2 | 0 | 0.34 |
Weiquan Yu | 3 | 0 | 0.34 |
Suiyu Zhang | 4 | 0 | 0.34 |
Jiaojiao Wang | 5 | 0 | 0.34 |
Dingguo Yu | 6 | 0 | 0.68 |