Research Article
Computer-Assisted Diagnosis for Diabetic Retinopathy Based on Fundus Images Using Deep Convolutional Neural Network
Table 5
Confusion matrix of the classification results from the two DCNN networks, before performing SVM classification.
| Determined level | Ground-truth level | Network 1 | Network 2 | 0 | 1 | 2 | 3 | 4 | 0 | 1 | 2 | 3 | 4 |
| 0 | 39031 | 2693 | 2361 | 75 | 160 | 38310 | 2233 | 1617 | 51 | 78 | 1 | 118 | 438 | 185 | 0 | 1 | 327 | 735 | 290 | 1 | 3 | 2 | 339 | 626 | 5058 | 738 | 392 | 790 | 786 | 5509 | 620 | 345 | 3 | 0 | 0 | 170 | 332 | 102 | 4 | 1 | 338 | 486 | 185 | 4 | 43 | 5 | 85 | 69 | 551 | 100 | 7 | 105 | 56 | 595 | Accuracy (%) | 98.74 | 11.64 | 64.36 | 27.35 | 45.69 | 96.91 | 19.54 | 70.01 | 40.03 | 49.34 | Overall accuracy (%) | 84.76 | 85.18 |
|
|