{"id":2124,"date":"2015-06-18T06:22:39","date_gmt":"2015-06-18T06:22:39","guid":{"rendered":"http:\/\/biomedpharmajournal.org\/?p=2124"},"modified":"2020-04-25T02:57:33","modified_gmt":"2020-04-25T02:57:33","slug":"computer-aided-screening-of-optic-disc-in-retinal-images-using-binary-orientation-map","status":"publish","type":"post","link":"https:\/\/biomedpharmajournal.org\/staging\/vol8no1\/computer-aided-screening-of-optic-disc-in-retinal-images-using-binary-orientation-map\/","title":{"rendered":"Computer Aided Screening of Optic Disc in Retinal Images Using Binary Orientation Map"},"content":{"rendered":"

Introduction<\/strong><\/p>\n

The detection of optic disc (OD) is a first step for the diagnosis of different pathologies of retinal diseases such as glaucoma, diabetic retinopathy. The optic disc is in a vertical oval shape with the horizontal and vertical average dimension of 1.76mm and 1.92mm respectively. It is located\u00a0 at\u00a0 the\u00a0 distance\u00a0 of\u00a0 3\u00a0 to\u00a0 4mm\u00a0 from\u00a0 the\u00a0 nasal\u00a0 side\u00a0 of fovea (Haseeb \u00a0Ahsan, 2015)1<\/sup>.The\u00a0 dimension\u00a0 of\u00a0 optic\u00a0 disc\u00a0 varies\u00a0 from\u00a0 person\u00a0 to person but its diameter remains the\u00a0 same (80-100 pixels) for standard fundus\u00a0 image (Peter H. Scanlon,2015)2<\/sup>.The optic disc is a bright yellowish disc\u00a0 that\u00a0 transmits\u00a0 electrical\u00a0 impulses\u00a0 from\u00a0 retina\u00a0 to\u00a0 brain where\u00a0 the\u00a0 occurrence\u00a0 of\u00a0 blood\u00a0 vessels and\u00a0 optic\u00a0 nerve\u00a0 fibre\u00a0 takes place (David et al<\/em>., 2015)3<\/sup> .The optic disc does not contain any receptors hence it is also called as blind spot. The OD vicinity consists of large vessels that are used for vessel tracking methods. For any\u00a0 retinal\u00a0 fundus\u00a0 image\u00a0 OD\u00a0 occupies\u00a0 about\u00a0 one\u00a0 seventh\u00a0 of the entire image.\u00a0 The health status of the human retina can be determined from the physical characteristics of OD (Ahmed et al<\/em>., 2015)4<\/sup>.<\/p>\n

There are many approaches (5-19)<\/sup> that have been proposed for the automatic detection of OD.\u00a0 The proposed OD detection methods are categorized as mathematical morphology (5-9)<\/sup>, segmentation algorithms (10-15)<\/sup>, Optimization(16-18)<\/sup> and template matching(19)<\/sup>. Mathematical morphology plays a major role in retinal image<\/p>\n

 <\/p>\n\n\n\n
\"figure<\/td>\nFigure 1: (a) Retinal image landmarks (b) OD localized\u00a0 image<\/strong><\/p>\n

Click here to view full figure<\/a><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

 <\/p>\n

processing that includes dilation, erosion, opening and closing. The different techniques that are implemented in mathematical morphology for the detection of optic disc are Principle\u00a0 Component\u00a0 Analysis (PCA) (Sinthanayothin et al<\/em>.,2010)5<\/sup>,\u00a0 morphological based\u00a0 edge\u00a0 detection (Arturo\u00a0 Aquino et al<\/em>., 2010)6<\/sup>,\u00a0 vessel\u00a0 direction\u00a0 matched\u00a0 filter\u00a0 method (Aliaa\u00a0 Abdel et al<\/em>.,2008)7<\/sup>, Sobel\u00a0 operator (Abdel-Ghafar \u00a0et al<\/em>.,2007)8<\/sup> and bit\u00a0 plane\u00a0 slicing(Ashok\u00a0 kumar et al<\/em> 2012)9<\/sup>.<\/p>\n

Segmentation algorithms uses to divides the fundus image into multiple segments in order to locate OD and its boundaries. This methods are vessel\u00a0 classification (Jeline 2010)10<\/sup>,\u00a0\u00a0 thresholding (F.\u00a0 ter\u00a0 Haar ,2005)11<\/sup>, \u00a0Snake method(Alireza\u00a0 Osareh \u00a0et al<\/em>., 2002)12<\/sup>, Active contour(Gopal\u00a0 Datt\u00a0 Joshi et al<\/em>.,2010)13<\/sup>, Deformation\u00a0 contour\u00a0 method\u00a0 (Juan\u00a0 Xu et al<\/em>.,2006)14<\/sup> and Wavelet transform(Pallawala et al<\/em>.,2004)15<\/sup>. Optimization\u00a0 technique\u00a0 are used to locate OD, these methods includes fuzzy\u00a0 convergence(Kenneth et al<\/em>.,2006)16<\/sup>,\u00a0 Forward\u00a0 selection\u00a0\u00a0 and\u00a0 backward elimination(Bowd et al<\/em>.,2002)17<\/sup> and KNN regressor (Niemeijer \u00a0et al<\/em>.,2009)18<\/sup>.The template based methods are also used to localize OD these method hausdorff template matching(Marc\u00a0 Lalonde \u00a0et al<\/em>.,2001)19<\/sup>.<\/p>\n

Material<\/strong><\/p>\n

Two publicly available datasets were used to test the proposed method. The STARE subset (20)<\/sup> contains 81 fundus images that were used for evaluating their automatic OD localization method. The images were captured using a Top Con TRV-50 fundus camera at 35o field-of-view (FOV), and subsequently digitized at 605 X 700, 24-bits pixel. The dataset contains 31 images of normal retinas and 50 of diseased retinas.<\/p>\n

The second dataset used is the DRIVE dataset (21)<\/sup>, established to facilitate comparative studies on retinal OD localization. The dataset consists of a total of 40 color fundus photographs used for making actual clinical diagnoses, where 33 photographs do not show any sign of diabetic retinopathy and seven show signs of mild early diabetic retinopathy. The \u201324 bits, 768 by 584 pixels color images are in compressed JPEG-format, and acquired using a Canon CR5 non-mydriatic 3CCD camera with a 45o<\/sup> FOV.<\/p>\n

Methodology<\/strong><\/p>\n

This section presents the proposed OD detection technique. In particular, we divide this section into four subsections, which deal with the retinal image pre-processing, designed line operator, OD detection, and discussion, respectively.<\/p>\n

Retinal Image Pre-processing<\/strong><\/p>\n

Retinal images need to be pre-processed before the OD detection. As the proposed technique makes use of the circular brightness structure of the OD, the lightness component of a retinal image is first extracted. We use the lightness component within the LAB color space, where the OD detection usually performs the best (Osareh et al<\/em>.,2002). For the retinal image in Fig. 1(a), Fig. 2(a) shows the corresponding lightness image.The retinal image is then smoothed to enhance the circular brightness structure associated with the OD. We use a bilateral smoothing filter (Tomasi et al<\/em>.,1998) that combines geometric closeness and photometric similarity as follows:<\/p>\n

h(x) = k(x) c(\u03be, x) s(f(\u03be); f(x)) d \u03be (1)<\/p>\n

With the normalization factor<\/p>\n

k(x) = s(f(\u03be); f(x)) d \u03be\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 (2)<\/p>\n

where f (x) denotes the retinal image under study. c(\u03be, x) and s(f(\u03be), f (x)) measure the geometric closeness and the photometric similarity between the neighborhood center x and a nearby point \u03be. We set both c(\u03be, x) and s(f(\u03be), f(x)) by Gaussian functions. The geometric spread \u03c3d and the photometric spread \u03c3r of the two Gaussian functions are typically set at 10 and 1 as reported in [23]. For the retinal image in Fig. 2(a), Fig. 2(b) shows the filtered retinal image.<\/p>\n\n\n\n
\"figure<\/td>\nFigure2: Retinal mage pre-processing. (a) Lightness of the example retinal image in LAB color space. (b) Enhanced retinal image by bilateral smoothing where multiple crosses along a circle label the pixels to be used to illustrate the image variation along multiple oriented line segments.<\/strong><\/p>\n

 <\/p>\n

Click here to view full figure<\/a><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

 <\/p>\n

Line operator<\/strong><\/p>\n

A line operator is designed to detect circular regions that have similar brightness structure as the OD. For each image pixel at (x, y), the line operator first determines n line segments Li<\/sub>, i = 1. . . n of specific length p (i.e., number of pixels) at multiple specific orientations that center at \u00a0\u00a0(x, y). The image intensity along all oriented line segments can thus be denoted by a matrix\u00a0\u00a0\u00a0\u00a0\u00a0 I(x,y)nxp, where each matrix row stores the intensity of p image pixels along one specific line segment. The line operator that uses 20 oriented line segments and sets the line length\u00a0\u00a0 p = 21, each line segment Li at one specific orientation can be divided into two line segments Li,1<\/sub> and Li,2<\/sub> of the same length (p \u2212 1)\/2 by the image pixel under study (i.e., the black cell). The image variation along the oriented line segments can be estimated as follows:<\/p>\n

Di<\/sub>(x,y) =\u2551 fmdn<\/sub> (Ili,1<\/sub>(x,y)) – fmdn<\/sub> (Ili,2<\/sub>(x,y)) \u2551, i=1\u2026n\u00a0\u00a0\u00a0\u00a0 \u00a0\u00a0\u00a0\u00a0(3)<\/p>\n

Where fmdn<\/sub> (\u00b7) denotes a median function.\u00a0 fmdn (Ili,1<\/sub> (x, y)) and fmdn (Ili,2<\/sub> (x, y)) return the median image intensity along Li,1<\/sub>and Li,2<\/sub>,respectively. D = [Di<\/sub>(x, y)..Di<\/sub> (x, y). . .Dn<\/sub> (x, y)] is, therefore, a vector of dimension n that stores the image variations along n-oriented line segments.<\/p>\n

The orientation of the line segment with the maximum\/minimum variation has specific pattern that can be used to locate the OD accurately. For retinal image pixels, which are far away from the OD, the orientation of the line segment with the maximum\/minimum variation is usually arbitrary, but for those around the OD, the image variation along Lc [labeled in Fig. 1(b)] usually reach the maximum, whereas that along Lt reaches the minimum. Fig. 4 shows the image variation vectors D(x, y) of eight pixels that are labeled by crosses along a circle shown in Fig. 2(b). Suppose that there is a Cartesian coordinate system centered at the OD, as shown in Fig. 2(b). For the retinal image pixels in quadrants I and III, the image variations along the 1st\u201310th [i.e., Lt<\/sub> in Fig. 1(b)] and the 11th\u201320th (i.e., Lc<\/sub> ) line segments labeled in Fig. 3 reach the minimum and the maximum, respectively, as shown in Fig. 4. But for the retinal image pixels in quadrants II and IV, the image variations along the 1st\u201310th and the 11th\u201320th line segments instead reach the maximum and the minimum, respectively.<\/p>\n

An orientation is constructed based on the orientation of the line segment with the maximum (or minimum) variation as follows:<\/p>\n

O (x,y) = argmax D(x,y)\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 \u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 (4)<\/p>\n

i<\/p>\n

where D(x,y) denotes the image variation vector evaluated in (3). In addition, a binary orientation map can also be constructed by classifying the orientation of the line segment with the maximum variation into two categories as follows:<\/p>\n

\"frmla-1\"<\/p>\n

 <\/p>\n

where n refers to the number of the oriented line segments used in the line operator<\/p>\n\n\n\n
\"figure<\/td>\nFigure 3: Orientation map of the retinal image in Fig. 2(b). (a) Gray orientation map that is determined by using (4). (b) Binary orientation map that is determined by using (5).<\/strong><\/p>\n

Click here to view full figure<\/a><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

 <\/p>\n

For the retinal image in Fig. 1(a), Fig. 3(a) and (b) shows the determined gray orientation map and binary orientation map, respectively. As shown in Fig. 3(a), for retinal image pixels in quadrants I and III around the OD, the orientation map is a bit dark because the orientation of the line segment with the maximum variation usually lies between 1 and (n\/2) + 1. However, for retinal image pixels in quadrants II and IV, the orientation map is bright because the orientation of the line segment with the maximum variation usually lies between n\/2 and n. The binary orientation map in Fig. 3(b) further verifies such orientation pattern. The OD will then be located by using the orientation map to be described in the following.<\/p>\n

Proposed circular convolution \u00a0<\/strong><\/p>\n

We use a line operator of 20 oriented line segments because line operators with more line segments have little effect on the orientation map. The line length p is set as follows:<\/p>\n

p = k R\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 \u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 \u00a0(6)<\/p>\n

where R denotes the radius of the central circular region of retinal images as illustrated in Fig. 1(a). Parameter k controls the line length, which usually lies between 1\/10 and 1\/5 based on the relative OD size within retinal images [25]. The use of R incorporates possible variations of the image resolution.<\/p>\n

The specific pattern within the orientation map is captured by a 2-D circular convolution mask shown at the upper left corner of two peak images in Fig. 4.b. As shown in Fig. 4, the convolution mask can be divided into four quadrants, where the cells within quadrants I and III are set at\u22121, whereas those within quadrants II and IV are set at 1 based on the specific pattern within the orientation map. An orientation map can thus be converted into a peak image as follows:<\/p>\n

\"frmla-2\"<\/p>\n

where (xo<\/sub> yo<\/sub> ) denotes the position of the retinal image pixel under study. M(x, y) and\u00a0\u00a0\u00a0 O(x, y) refer to the value of the convolution mask and the orientation map at (x, y), respectively. Parameter m denotes the radius of the circular convolution mask that can be similarly set as p.<\/p>\n\n\n\n
\"figure<\/td>\nFigure 4: Peak images determined by a 2-D circular convolution mask shown in the upper left corner. (a) Peak image produced through the convolution of the gray orientation map in Fig. 3(a). (b) Peak image produced through the convolution of the binary orientation map in Fig. 3(b).<\/strong><\/p>\n

Click here to view full figure<\/a><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

For the orientation maps in Fig. 5(a) and (b), Fig. 6(a) and (b) shows the determined peak images. As shown in Fig. 4, a peak is properly produced at the OD position. On the other hand, a peak is also produced at the macula center (i.e., fovea) that often has similar peak amplitude to the peak at the OD center. This can be explained by similar brightness variation structure around the macula, where the image variation along the line segment crossing the macula center reaches the maximum, whereas that along the orthogonal line segment [similar to Lc and Lt in Fig. 1(b)] reaches the minimum. The only difference is that the OD center is brighter than the surrounding pixels, whereas the macula center is darker.<\/p>\n

We, therefore, first classify the peaks into an OD category and a macula category, respectively. The classification is based on the image difference between the retinal image pixels at the peak center and those surrounding the peak center. The image difference is evaluated by two concentric circles as follows:<\/p>\n

\"frmla-3\"<\/p>\n

where I refers to the retinal image under study and d denotes the distance between the peak and the surrounding retinal image pixels. R1 and R2 specify the radius of an inner concentric circle and an outer concentric circle where R2 is set at 2 R1. Ni and No denote the numbers of the retinal image pixels within the two concentric circles. In our system, we set R1at\u00a0\u00a0\u00a0 (p \u2212 1)\/2, where p is the length of the line operator. The peak can, therefore, be classified to the OD or macula category, if the image difference is positive or negative, respectively. Finally, we detect the OD by a score that combines both the peak amplitude and the image intensity difference that by itself is also a strong indicator of the OD<\/p>\n

S(x,y) = P(x,y) (Diff(x,y) * (Diff(x,y) > 0))\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 (9)<\/p>\n

Where P(x,y) denotes the normalized peak image. The symbol * denotes dot product and Diff(x, y) > 0 sets all retinal image pixels with a negative image difference to zero. The OD can, therefore, be detected by the peak in the OD category that has the maximum score. For the example retinal image in Fig. 1(a), Fig. 5(a) shows the score image determined by the peak image in Fig. 4(b). It should be noted that the image difference is evaluated only at the detected peaks in practical implementation. The score image in Fig. 5(a) (as well as in Fig. 5(b), 9, and 10) where the image difference is evaluated at every pixel is just for the illustration purpose<\/p>\n\n\n\n
\"figure<\/td>\nFigure 5: OD detection. Score image by (9) for OD detection<\/strong><\/p>\n

Click here to view full figure<\/a><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

 <\/p>\n

Results And Discussion<\/strong><\/p>\n

The proposed method was implemented in MATLAB where runs needed on average 1.5min for each image on a laptop (Pentium(R) Dual-Core2 CPU and T4300@ 2.10GHz, 3.00 GB RAM 64-bit OS). The step by step output is shown in fig.6.(a)-(h)<\/p>\n

The proposed method was evaluated both normal and abnormal retinal images using the subset of the DRIVE, STARE, and the success percentage was found to be 98.34% for all datasets(shown in table I).<\/p>\n

 <\/p>\n\n\n\n
\"figure<\/td>\nFigure 6: (a)Input Image, (b)LAB – L Component, (c)Smooth Image, (d)Orientation map (e) Binary orientation map (f) Circular Convolution(g)Peak circular convolution mask(h) Detected Optic Disc<\/strong><\/p>\n

Click here to view full figure<\/a><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

 <\/p>\n

Table 1: Results<\/strong><\/p>\n\n\n\n\n\n\n
Dataset<\/strong><\/td>\nNo of Images tested<\/strong><\/td>\nNo of images failed<\/strong><\/td>\nSuccess Percentage (%)<\/strong><\/td>\n<\/tr>\n
DRIVE<\/td>\n40<\/td>\n0<\/td>\n100<\/td>\n<\/tr>\n
STARE<\/td>\n81<\/td>\n2<\/td>\n97.53<\/td>\n<\/tr>\n
Total<\/td>\n121<\/td>\n2<\/td>\n98.34<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

The implemented binary orientation map is tested with our proposed method for all the 121 images of STARE and DRIVE datasets respectively it is achieve more comprehensive result when compare to other methods. Both data sets results are compared with other state of art methods (Fig.7a and b).<\/p>\n

 <\/p>\n\n\n\n
\"figure<\/td>\nFigure 7: Comparison of the OD detection methods (a) STARE dataset (b) DRIVE dataset<\/strong><\/p>\n

Click here to view full figure<\/a><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

 <\/p>\n

Conclusion And Future Work<\/strong><\/p>\n

This paper presented a simple method for OD detection using binary orientation map. The proposed approach achieved better results as reported. An extension for this study could be as follows. region of interest would be the main objective of future work. Optic disc can be detected easily after segmentation. A myriad of Optic Disc detection methods are available.The image segmentation is the foundation for the retinal fundus images.\u00a0 Efficient algorithms for segmenting the<\/p>\n

The method which is compatible with the segmentation algorithm should be used. Diseases like Glaucoma and Diabetic Retinopathy are in rise and Optic Disc is an important indicator of these diseases. Hence the new methodologies to detect Optic Disc and the efficient use of already existing methods are the interest of future work.<\/p>\n

Reference<\/strong><\/p>\n

    \n
  1. Haseeb Ahsan, \u201dDiabetic retinopathy\u00a0\u2013 Biomolecules and multiple pathophysiology\u201d Diabetes & Metabolic Syndrome: Clinical Research & Reviews,\u00a0Vol 9( 1),\u00a0<\/strong>pp 51-54, 2015.<\/li>\n
  2. Peter H. Scanlon,\u201dDiabetic retinopathy\u201d, Medicine,\u00a0Vol 43(1),pp 13-19, 2015.<\/li>\n
  3. David S.M. Burton, Zia I. Carrim \u201cAngiographic Evidence of Peripheral Ischemia in\u00a0Diabetic Retinopathy\u00a0and the Risk of Impending Neovascularisation\u201d Canadian Journal of Diabetes,\u00a0Vol 39 (1), pp 14-15, 2015.<\/li>\n
  4. Ahmed M. Abu El-Asrar, Gert De Hertogh, Kathleen van den Eynde, Kaiser Alam, Katrien Van Raemdonck, Ghislain Opdenakker, Jo Van Damme, Karel Geboes, Sofie Struyf,\u201dMyofibroblasts in proliferative\u00a0diabetic retinopathy\u00a0can originate from infiltrating fibrocytes and through endothelial-to-mesenchymal transition (EndoMT)\u201d, Experimental Eye Research,\u00a0Vol 132(1), pp 179-189,2015.<\/li>\n
  5. Sinthanayothin,J. F.\u00a0 Boyce,\u00a0 H.\u00a0 L.\u00a0 Cook,\u00a0 and\u00a0 T.H.Williamson.,\u201dAutomated\u00a0 localization\u00a0 of\u00a0 Optic\u00a0 Disc,fovea,and retinal\u00a0 blood\u00a0 vessels\u00a0 from\u00a0 digital\u00a0 fundus\u00a0 images\u201d.,\u00a0 Br\u00a0 J Ophthalmol. 1999;83(8):902-10.<\/li>\n
  6. Arturo Aquino,\u00a0 Manuel\u00a0 Emilio\u00a0 Geg\u00fandez-Arias,\u00a0 and\u00a0 Diego Mar\u00edn .,\u201dDetecting the optic disc boundary in digital fundus images using\u00a0 morphological,\u00a0 edge\u00a0 detection,\u00a0 and\u00a0 feature\u00a0 extraction techniques\u201d.,IEEE\u00a0 \u00a0 Med.\u00a0 Imag.,\u00a0\u00a0 \u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0Nov\u00a0 2010.,vol.\u00a0 29,\u00a0 pp. 1860\u20131869.<\/li>\n
  7. Aliaa Abdel-Haleim\u00a0 Abdel-Razik\u00a0 Youssif,\u00a0 Atef\u00a0 Zaki\u00a0 Ghalwash, and\u00a0 Amr\u00a0 Ahmed\u00a0 Sabry\u00a0 Abdel-Rahman\u00a0 Ghoneim*.,\u201d\u00a0 Optic\u00a0 Disc detection\u00a0 from\u00a0 normalized\u00a0 digital\u00a0 fundus\u00a0 images\u00a0 by\u00a0 means\u00a0 of\u00a0 a vessels\u2019\u00a0 direction\u00a0 matched\u00a0 filter\u201d, IEEE\u00a0 \u00a0 Med. Imag.2008.,vol.27.,pp.11-18.<\/li>\n
  8. A.\u00a0 Abdel-Ghafar,\u00a0 T.\u00a0 Morris,\u00a0 T.\u00a0 Ritchings,\u00a0 I.\u00a0 Wood.,\u201dProgress towards automated detection and characterization of the\u00a0 optic disc in glaucoma and diabetic retinopathy\u201d., Med Inform Internet Med. 2007 ;32 (1):19-25.<\/li>\n
  9. Ashok\u00a0 kumar,S.Priya,\u00a0 Varghese\u00a0 Paul.,\u201d\u00a0 Automatic\u00a0 Feature Detection\u00a0 in\u00a0 Human\u00a0 Retinal\u00a0 Imagery\u00a0 using\u00a0 Bitplane\u00a0 Slicing\u00a0 and Mathematical\u00a0 Morphology\u201d.,\u00a0 European\u00a0 Journal\u00a0 of\u00a0 Scientific Research, Vol.80\u00a0 No.1\u00a0 (2012),\u00a0 pp.57-67<\/li>\n
  10. F. Jeline, C. Depardieu, C. Lucas, D.J. Cornforth, W. Huang and M.J. Cree.,\u201d Towards Vessel Characterisationin the Vicinity of the Optic Disc in Digital Retinal images\u201d.,<\/li>\n
  11. ter\u00a0 Haar,\u201cAutomatic\u00a0 localization\u00a0 of\u00a0 the\u00a0 optic\u00a0 disc\u00a0 in\u00a0 digital colour\u00a0 images\u00a0 of\u00a0 the\u00a0 human\u00a0 retina,\u201d M.S.\u00a0 thesis,\u00a0 Utrecht University, Utrecht,The Netherlands, 2005.<\/li>\n
  12. Alireza Osareh,\u00a0 Majid\u00a0 Mirmehdi,\u00a0 Barry\u00a0 Thomas\u00a0 and\u00a0 Richard Markham.,\u201cComparison of colour spaces for optic disc localisation in\u00a0 retinal\u00a0 images\u201d.,\u00a0 Proceedings\u00a0 of\u00a0 the\u00a0 16th\u00a0 International Conference on Pattern Recognition . Suen, (eds.), pp. 743\u2013746. August 2002.<\/li>\n
  13. Gopal Datt\u00a0 Joshi,\u00a0 Rohit\u00a0 Gautam,\u00a0 Jayanthi\u00a0 Sivaswamy,\u00a0 \u00a0 R. Krishnadas.,\u00a0 \u201cRobust\u00a0 Optic\u00a0 Disk\u00a0 Segmentation\u00a0 from\u00a0 Colour Retinal Images\u201d .,\u00a0\u00a0 Proceedings of the Seventh Indian Conference on\u00a0 Computer\u00a0 Vision,\u00a0 Graphics and\u00a0 Image\u00a0 Processing, \u00a0pp.330-336 ACM New York, NY, USA \u00a92010.<\/li>\n
  14. Juan Xu,\u00a0 Opas\u00a0 Chutatape,\u00a0 Eric\u00a0 Sung,Ce\u00a0 Zheng,\u00a0 Paul\u00a0 Chew\u00a0 Tec Kuan.,\u00a0 \u201cOptic\u00a0 disk\u00a0 feature\u00a0 extraction\u00a0 via\u00a0 modified\u00a0 deformable model\u00a0 technique\u00a0 for\u00a0 glaucoma\u00a0 analysis\u201d,\u00a0 Pattern Recognition, 2006<\/li>\n
  15. Pallawala,Wynne Mong\u00a0 Li\u00a0 Lee, Guan\u00a0 Au\u00a0 Eong, \u201cAutomated Optic Disc Localization and Contour Detection\u00a0 Using Ellipse\u00a0 Fitting\u00a0 and\u00a0 Wavelet\u00a0 Transform\u201d,\u00a0 Lecture\u00a0 Notes\u00a0 in Computer Science Vol. 3022 pp.139-151. , 2004,<\/li>\n
  16. Kenneth \u00a0 Tobin,\u00a0 Edward\u00a0 Chaum,\u00a0 V. \u00a0Priya\u00a0 Govindasamya, Thomas\u00a0 P.\u00a0 Karnowskia,\u00a0 Omer\u00a0 Sezer., \u201cCharacterization\u00a0 of\u00a0 the optic\u00a0 disc\u00a0 in\u00a0 retinal\u00a0 imagery\u00a0 using\u00a0 a probabilistic\u00a0 approach\u201d Vol 82006;10pages; \u00a0DOI: 10.1117\/12.641670.<\/li>\n
  17. Bowd C, Chan\u00a0 K, Zangwill\u00a0 LM, Goldbaum\u00a0 MH, Lee TW, Sejnowski\u00a0 TJ, Weinreb\u00a0 ,\u00a0 \u201cComparing\u00a0 neural\u00a0 networks and\u00a0 linear\u00a0 discriminant\u00a0 functions\u00a0 for\u00a0 glaucoma\u00a0 detection\u00a0 using\u00a0 confocal scanning laser ophthalmoscopy of the optic disc\u201d Invest Ophthalmol Vis Sci.Vol. 43(11),pp3444-54,2002.<\/li>\n
  18. Niemeijer M, Abr\u00e0moff MD, van Ginneken B., \u201cFast Detection of the Optic\u00a0 Disc\u00a0 and\u00a0 Fovea\u00a0 in\u00a0 Color\u00a0 Fundus\u00a0 Photographs\u201d.,Med Image Anal. vol.13(6),pp859\u2013870, 2009.<\/li>\n
  19. Marc Lalonde,\u00a0 Mario\u00a0 Beaulieu,\u00a0 and\u00a0 Langis\u00a0 ,\u00a0 \u201cFast\u00a0 and Robust Optic Disc Detection Using Pyramidal Decomposition and Hausdorff-Based\u00a0 Template\u00a0 Matching\u201d.,\u00a0 IEEE\u00a0 Trans.\u00a0 Med.\u00a0 Imag., 2001.,vol. 20, pp. 1193-1200.<\/li>\n
  20. STARE Project Website Clemson Univ., Clemson, SC [Online]. Available: http:\/\/www.ces.clemson.edu~ahoover\/stare<\/li>\n
  21. Research section, digital retinal image for vessel extraction (DRIVE) database University Medical Center Utrecht, Image Sciences Institute,Utrecht, The Netherlands [Online]. Available: http:\/\/www.isi.uu.nl\/Research\/Databases\/DRIV<\/li>\n
  22. Osareh, M. Mirmehdi, B. Thomas, and R. Markham, \u201cComparison of colour spaces for optic disc localisation in retinal images,\u201d in Proc. Int. Conf. Pattern Recognit., vol. 1, 2002, pp. 743\u2013746.<\/li>\n
  23. Tomasi and R. Manduchi, \u201cBilateral Filtering for Gray and Color Images,\u201d in Proc. IEEE Int. Conf. Comp. Vis., 1998, pp. 839\u2013846.<\/li>\n
  24. Mahfouz AE,\u00a0 Fahmy\u00a0 \u00a0 \u201cFast\u00a0 localization\u00a0 of\u00a0 the\u00a0 optic\u00a0 disc\u00a0 using \u00a0projection\u00a0 of image\u00a0 features\u201d\u00a0 IEEE\u00a0 Transactions\u00a0 on\u00a0 Image\u00a0 Processing , 2010;19(12):3285\u20139.<\/li>\n
  25. Lu \u00a0 \u201cAccurate\u00a0 and\u00a0 efficient\u00a0 optic\u00a0 disc\u00a0 detection\u00a0 and\u00a0 segmentation\u00a0 by\u00a0 a\u00a0 circular transformation\u201d,\u00a0 IEEE\u00a0 Transactions\u00a0 on\u00a0 Medical\u00a0 Imaging , 2011;30(12):2126\u201333.<\/li>\n
  26. Amin Dehghani, Hamid Abrishami Moghaddam,Mohammad-Shahram Moin,\u201d Optic disc localization in retinal images using histogram matching\u201d, EURASIP Journal on Image and Video Processing, vol.1(19),pp.1-11,2012.<\/li>\n
  27. Amin Dehghani, Mohammad-Shahram Moin and Mohammad Saghafi,\u201d Localization of the Optic Disc Center in Retinal Images Based on the Harris Corner Detector\u201d, Biomed Eng Lett Vol.2,pp.198-206,2012.<\/li>\n
  28. Kavitha, Duraiswamy, \u201cAn automatic retinal optic disc segmentation algorithm using differential windowing in the polar coordinate domain\u201d, Australian Journal of Electrical Electronics Engineering, Vol. 10(1), pp.33-44,2013.<\/li>\n
  29. Avinash Ramakanth , R. Venkatesh Babu, “, Approximate Nearest Neighbour Field based Optic Disk Detection”, Computerized Medical Imaging and Graphics,Vol 38(1),pp 49\u201356,2014.<\/li>\n
  30. Neelam Sinha, Venkatesh Babu,\u201d optic disk localization using l1 minimization\u201d, IEEE International Conference on Image Processing, DOI: 10.1109\/ICIP.2012.6467488,2012.<\/li>\n<\/ol>\n

     <\/p>\n

     <\/p>\n","protected":false},"excerpt":{"rendered":"

    Introduction The detection of optic disc (OD) is a first […]<\/p>\n","protected":false},"author":3,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[3],"tags":[],"class_list":["post-2124","post","type-post","status-publish","format-standard","hentry","category-vol8no1"],"_links":{"self":[{"href":"https:\/\/biomedpharmajournal.org\/staging\/wp-json\/wp\/v2\/posts\/2124","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/biomedpharmajournal.org\/staging\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/biomedpharmajournal.org\/staging\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/biomedpharmajournal.org\/staging\/wp-json\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/biomedpharmajournal.org\/staging\/wp-json\/wp\/v2\/comments?post=2124"}],"version-history":[{"count":5,"href":"https:\/\/biomedpharmajournal.org\/staging\/wp-json\/wp\/v2\/posts\/2124\/revisions"}],"predecessor-version":[{"id":32832,"href":"https:\/\/biomedpharmajournal.org\/staging\/wp-json\/wp\/v2\/posts\/2124\/revisions\/32832"}],"wp:attachment":[{"href":"https:\/\/biomedpharmajournal.org\/staging\/wp-json\/wp\/v2\/media?parent=2124"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/biomedpharmajournal.org\/staging\/wp-json\/wp\/v2\/categories?post=2124"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/biomedpharmajournal.org\/staging\/wp-json\/wp\/v2\/tags?post=2124"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}