Table 1 Summary of the state-of-the-art methods for DME detection
From: Machine learning techniques for diabetic macular edema (DME) classification on SD-OCT images
References | Diseases | Data size | Pre-processing | Features | Representation | Classifier | Evaluation | Results | |||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
AMD | DME | Normal | De-noise | Flatten | Aligning | Cropping | |||||||
Srinivansan et al. [10] | \(\checkmark \) | \(\checkmark \) | \(\checkmark \) | 45 | \(\checkmark \) | \(\checkmark \) | \(\checkmark \) | HoG | Linear-SVM | ACC | 86.7%,100%,100% | ||
Venhuizen et al. [11] | \(\checkmark \) | \(\checkmark \) | 384 | Texton | BoW, PCA | RF | AUC | 0.984 | |||||
Liu et al. [12] | \(\checkmark \) | \(\checkmark \) | \(\checkmark \) | 326 | \(\checkmark \) | \(\checkmark \) | Edge, LBP | PCA | SVM-RBF | AUC | 0.93 | ||
Lemaître et al. [13] | \(\checkmark \) | \(\checkmark \) | 32 | \(\checkmark \) | LBP, LBP-TOP | PCA, BoW, Histogram | RF | SE, SP | 87.5%, 75% | ||||
Sankar et al. [15] | \(\checkmark \) | \(\checkmark \) | 32 | \(\checkmark \) | \(\checkmark \) | \(\checkmark \) | Pixel-intensities | PCA | Mahalanobis-distance to GMM | SE, SP | 80%, 93% | ||
Albarrak et al. [16] | \(\checkmark \) | \(\checkmark \) | 140 | \(\checkmark \) | \(\checkmark \) | LBP-TOP, HoG | PCA | Bayesian network | SE, SP | 92.4%, 90.5% | |||