Diabetic Retinopathy Detection Using Local Extrema Quantized Haralick Features with Long Short-Term Memory Network

Table 3

Comparisons with other methods.

Authors	Method	Accuracy of classification	Sensitivity	Specificity	Positive predictive value (PPV)
[22] Muhammad Faisal et al.	Support vector machınes (SVMs)	Not reported	96.9%	100%	100%
[23] R. Radha et al.	Morphologıcal process and clusterıng technıque	98%	Not reported	Not reported	Not reported
[24] Sumandeep Kaur et al.	K-means colour compression and fuzzy logic	96%	94.7%	Not reported	Not reported
[25] Akara Sopharak et al.	Using fuzzy C-means clustering	87.28%	99.24%	42.77%	24.26
[2] Acharya et al.	Blood vessel, exudates, microaneurysms, hemorrhages	86%	82%	86%	Not reported
[26] Vujosevic et al.	Single lesions	Not reported	82%	92%	Not reported
[27] R.H.N.G. Ranamuka et al.	Fuzzy logic	Not reported	75.43%	99.99%,	Not reported
[28] Pavle et al.	Deep neural networks and anatomical landmark	Not reported	78%	Not reported	78%
[14] T. Walter et al.	Means of morphological reconstruction techniques	92.8%	Not reported	Not reported	92.4%
[29] E. Imani et al.	Signal separation algorithm	89.01%	99.93%	82.64%	Not reported
[30] Abdullah Saeed et al.	Digital analysis and mathematical morphology operations	86%	80%	84.69%	Not reported
Proposed method	Local extrema quantized Haralick features with Long Short-Term Memory (LSTM) network	95.45%	91.65%	95.45%	99.34%