From: Microaneurysm detection in fundus images using a two-step convolutional neural network
Free-response receiver operating characteristic results on E-Ophtha-MA dataset at average number of False positives per image | ||||||||
---|---|---|---|---|---|---|---|---|
Method | ||||||||
Sensitivity | ||||||||
FPs/Img | 1/8 | 1/4 | 1/2 | 1 | 2 | 4 | 8 | Classification method |
 Proposed method | 0.091 | 0.258 | 0.401 | 0.534 | 0.579 | 0.667 | 0.771 | CNN |
 Dashtbozorg [38] | 0.358 | 0.417 | 0.471 | 0.522 | 0.558 | 0.605 | 0.638 | RUSBoost |
 Chudzika [29] | 0.151 | 0.264 | 0.376 | 0.468 | 0.542 | 0.595 | 0.621 | CNN |
 Wu’s [39] | 0.063 | 0.117 | 0.172 | 0.245 | 0.323 | 0.417 | 0.573 | KNN |