Retinal image analysis is increasingly prominent as a non-intrusive diagnosis method in modern ophthalmology. In this paper, we present a novel method to segment blood vessels and optic disc in the fundus retinal images. The method could be used to support non-intrusive diagnosis in modern ophthalmology since the morphology of the blood vessel and the optic disc is an important indicator for diseases like diabetic retinopathy, glaucoma and hypertension. Our method takes as first step the extraction of the retina vascular tree using the graph cut technique. The blood vessel information is then used to estimate the location of the optic disc. The optic disc segmentation is performed using two alternative methods. The Markov Random Field (MRF) image reconstruction method segments the optic disc by removing vessels from the optic disc region and the Compensation Factor method segments the optic disc using prior local intensity knowledge of the vessels. The proposed method is tested on three public data sets, DIARETDB1, DRIVE and STARE. The results and comparison with alternative methods show that our method achieved exceptional performance in segmenting the blood vessel and optic disc.
Performance analysis of retinal image blood vessel segmentationacijjournal
The retinal image diagnosis
is an important methodology for diabetic retinopathy detection and analysis. in
this paper, the morphological operations and svm classifier are used to detect and segment the blood
vessels from the retinal image. the proposed system consists of three stage
s
-
first is preprocessing of retinal
image to separate the green channel and second stage is retinal image enhancement and third stage is
blood vessel segmentation using morphological operations and svm classifier. the performance of the
proposed system is
analyzed using publicly available dataset
An Automated Systems for the Detection of Macular Ischemia based-Diabetic Ret...iosrjce
The proposed methodology in this paper marks out application for automatic detection of eye
diseases called Macular Ischemia using image processing techniques. In semi urban and rural areas large
percentages of people suffer from various eye diseases. For diagnoses of various eye diseases, Image processing
technique is used. . Diseases occur in Macula from retinal images have a huge type of textures, shapes and at
times they are difficult to be recognised and identified by doctors. Thus we are trying to optimize and develop
such system which is based on smart image recognition/classification algorithms. This proposed system
provides accuracy, uniformity and speed in performance and a high credence coefficient in results interpreting.
Keywords: Macular Ischemia, diagnosis, textures, consistence
An Approach for the Detection of Vascular Abnormalities in Diabetic Retinopathyijdmtaiir
Diabetic Retinopathy is a common complication of
diabetes that is caused by changes in the blood vessels of the
retina. The blood vessels in the retina get altered. Exudates are
secreted, micro-aneurysms and hemorrhages occur in the
retina. The appearance of these features represents the degree
of severity of the disease. In this paper the proposed approach
detects the presence of abnormalities in the retina using image
processing techniques by applying morphological processing
techniques to the fundus images to extract features such as
blood vessels, micro aneurysms and exudates. These features
are used for the detection of severity of Diabetic Retinopathy.
It can quickly process a large number of fundus images
obtained from mass screening to help reduce the cost, increase
productivity and efficiency for ophthalmologists.
International Journal of Computational Engineering Research(IJCER) is an intentional online Journal in English monthly publishing journal. This Journal publish original research work that contributes significantly to further the scientific knowledge in engineering and Technology.
Optic Disc and Macula Localization from Retinal Optical Coherence Tomography ...IJECEIAES
This research used images from Optical Coherence Tomography (OCT) examination as well as fundus images to localize the optical disc and macular layer of retina. The researchers utilized the OCT and fundus image to interpret the distance between macular center and optic disc in the image. The distance will express the area of macula that can be employed for further research. This distance could recognize the thickness of macula parameters diameter that will be used in localizing process of optic disc and macula. The parameters are the circle radius, the size of window’s filter, the constant value and the size of optic disc element structure as well as the size of macula. The results of this study are expected to improve the accuracy of macula detection that experience the edema.
Performance analysis of retinal image blood vessel segmentationacijjournal
The retinal image diagnosis
is an important methodology for diabetic retinopathy detection and analysis. in
this paper, the morphological operations and svm classifier are used to detect and segment the blood
vessels from the retinal image. the proposed system consists of three stage
s
-
first is preprocessing of retinal
image to separate the green channel and second stage is retinal image enhancement and third stage is
blood vessel segmentation using morphological operations and svm classifier. the performance of the
proposed system is
analyzed using publicly available dataset
An Automated Systems for the Detection of Macular Ischemia based-Diabetic Ret...iosrjce
The proposed methodology in this paper marks out application for automatic detection of eye
diseases called Macular Ischemia using image processing techniques. In semi urban and rural areas large
percentages of people suffer from various eye diseases. For diagnoses of various eye diseases, Image processing
technique is used. . Diseases occur in Macula from retinal images have a huge type of textures, shapes and at
times they are difficult to be recognised and identified by doctors. Thus we are trying to optimize and develop
such system which is based on smart image recognition/classification algorithms. This proposed system
provides accuracy, uniformity and speed in performance and a high credence coefficient in results interpreting.
Keywords: Macular Ischemia, diagnosis, textures, consistence
An Approach for the Detection of Vascular Abnormalities in Diabetic Retinopathyijdmtaiir
Diabetic Retinopathy is a common complication of
diabetes that is caused by changes in the blood vessels of the
retina. The blood vessels in the retina get altered. Exudates are
secreted, micro-aneurysms and hemorrhages occur in the
retina. The appearance of these features represents the degree
of severity of the disease. In this paper the proposed approach
detects the presence of abnormalities in the retina using image
processing techniques by applying morphological processing
techniques to the fundus images to extract features such as
blood vessels, micro aneurysms and exudates. These features
are used for the detection of severity of Diabetic Retinopathy.
It can quickly process a large number of fundus images
obtained from mass screening to help reduce the cost, increase
productivity and efficiency for ophthalmologists.
International Journal of Computational Engineering Research(IJCER) is an intentional online Journal in English monthly publishing journal. This Journal publish original research work that contributes significantly to further the scientific knowledge in engineering and Technology.
Optic Disc and Macula Localization from Retinal Optical Coherence Tomography ...IJECEIAES
This research used images from Optical Coherence Tomography (OCT) examination as well as fundus images to localize the optical disc and macular layer of retina. The researchers utilized the OCT and fundus image to interpret the distance between macular center and optic disc in the image. The distance will express the area of macula that can be employed for further research. This distance could recognize the thickness of macula parameters diameter that will be used in localizing process of optic disc and macula. The parameters are the circle radius, the size of window’s filter, the constant value and the size of optic disc element structure as well as the size of macula. The results of this study are expected to improve the accuracy of macula detection that experience the edema.
The main cause of eye diseases in the working human is Diabetic retinopathy. Eye disease can
be prevented if detects early. The extraction of blood vessels from retinal images is an essential and challenging
task in medical diagnosis and analysis. This paper describes the effective and efficient extraction of blood
vessels from retinal image by using Kirsch’s templates. The Kirsch’s edge operators detect the edges using eight
filters, generated by the compass rotation mechanism. The method is used to automatic detection of landmark
features of the fundus, such as the optic disc, fovea and blood vessels.
Retinal image analysis using morphological process and clustering techniquesipij
This paper proposes a method for the Retinal image analysis through efficient detection of exudates and
recognizes the retina to be normal or abnormal. The contrast image is enhanced by curvelet transform.
Hence, morphology operators are applied to the enhanced image in order to find the retinal image ridges.
A simple thresholding method along with opening and closing operation indicates the remained ridges
belonging to vessels. The clustering method is used for effective detection of exudates of eye. Experimental
result proves that the blood vessels and exudates can be effectively detected by applying this method on the
retinal images. Fundus images of the retina were collected from a reputed eye clinic and 110 images were
trained and tested in order to extract the exudates and blood vessels. In this system we use the Probabilistic
Neural Network (PNN) for training and testing the pre-processed images. The results showed the retina is
normal or abnormal thereby analyzing the retinal image efficiently. There is 98% accuracy in the detection
of the exudates in the retina .
Teamed with 2 students to research and implement the automation of diagnosis of Diabetic Retinopathy and co-ordinated with an Ophthalmologist to verify our implementation.
Responsibilities included MATLAB coding, algorithm testing, and product documentation.
• Automation in MATLAB involving retinal image analysis to help
Ophthalmologist increase the productivity and efficiency in a clinical
environment.
• Used Image Processing concepts such as Hough Transform, Bottom Hat
Transform, Edge Detection Technique and Morphological Operators.
Provided our algorithm and documentation to our research faculty advisor to enable him to continue this research to the next phase.
IMAGE PROCESSING Projects for M. Tech, IMAGE PROCESSING Projects in Vijayanagar, IMAGE PROCESSING Projects in Bangalore, M. Tech Projects in Vijayanagar, M. Tech Projects in Bangalore, IMAGE PROCESSING IEEE projects in Bangalore, IEEE 2015 IMAGE PROCESSING Projects, MATLAB Image Processing Projects, MATLAB Image Processing Projects in Bangalore, MATLAB Image Processing Projects in Vijayangar
Segmentation of the Blood Vessel and Optic Disc in Retinal Images Using EM Al...iosrjce
IOSR Journal of Computer Engineering (IOSR-JCE) is a double blind peer reviewed International Journal that provides rapid publication (within a month) of articles in all areas of computer engineering and its applications. The journal welcomes publications of high quality papers on theoretical developments and practical applications in computer technology. Original research papers, state-of-the-art reviews, and high quality technical notes are invited for publications.
C LASSIFICATION O F D IABETES R ETINA I MAGES U SING B LOOD V ESSEL A REASIJCI JOURNAL
Retina images are obtained from the fundus camera a
nd graded by skilled professionals. However there i
s
considerable shortage of expert observers has encou
raged computer assisted monitoring. Evaluation of
blood vessels network plays an important task in a
variety of medical diagnosis. Manifestations of
numerous vascular disorders, such as diabetic retin
opathy, depend on detection of the blood vessels
network. In this work the fundus RGB image is used
for obtaining the traces of blood vessels and areas
of
blood vessels are used for detection of Diabetic Re
tinopathy (DR). The algorithm developed uses
morphological operation to extract blood vessels. M
ainly two steps are used: firstly enhancement opera
tion
is applied to original retina image to remove noise
and increase contrast of retinal blood vessels. Se
condly
morphology operations are used to take out blood ve
ssels. Experiments are conducted on publicly availa
ble
DIARETDB1 database. Experimental results obtained b
y using gray-scale images have been presented.
Automated histopathological image analysis: a review on ROI extractioniosrjce
IOSR Journal of Computer Engineering (IOSR-JCE) is a double blind peer reviewed International Journal that provides rapid publication (within a month) of articles in all areas of computer engineering and its applications. The journal welcomes publications of high quality papers on theoretical developments and practical applications in computer technology. Original research papers, state-of-the-art reviews, and high quality technical notes are invited for publications.
Review of methods for diabetic retinopathy detection and severity classificationeSAT Journals
Abstract Diabetic Retinopathy is a serious vascular disorder that might lead to complete blindness. Therefore, the early detection and the treatment are necessary to prevent major vision loss. Though the Manual screening methods are available, they are time consuming and inefficient on a large image database of patients. Moreover, it demands skilled professionals for the diagnosis. Automatic Diabetic Retinopathy diagnosis systems can replace manual methods as they can significantly reduce the manual labor involved in the screening process. Screening conducted over a larger population can become efficient if the system can separate normal and abnormal cases, instead of the manual examination of all images. Therefore, Automatic Retinopathy detection systems have attracted large popularity in the recent times. Automatic retinopathy detection systems employ image processing and computer vision techniques to detect different anomalies associated with retinopathy. This paper reviews various methods of diabetic retinopathy detection and classification into different stages based on severity levels and also, various image databases used for the research purpose are discussed. Keywords— Automatic Diabetic Retinopathy detection, computer vision, Diabetic Retinopathy, image databases, image processing, manual screening
Automatic detection of optic disc and blood vessels from retinal images using...eSAT Publishing House
IJRET : International Journal of Research in Engineering and Technology is an international peer reviewed, online journal published by eSAT Publishing House for the enhancement of research in various disciplines of Engineering and Technology. The aim and scope of the journal is to provide an academic medium and an important reference for the advancement and dissemination of research results that support high-level learning, teaching and research in the fields of Engineering and Technology. We bring together Scientists, Academician, Field Engineers, Scholars and Students of related fields of Engineering and Technology
FUZZY CLUSTERING BASED GLAUCOMA DETECTION USING THE CDR sipij
Glaucoma is a serious eye disease, overtime it will result in gradual blindness. Early detection of thedisease will help prevent against developing a more serious condition. A vertical cup-to-disc ratio which isthe ratio of the vertical diameter of the optic cup to that of the optic disc, of the fundus eye image is an important clinical indicator for glaucoma diagnosis. This paper presents an automated method for the extraction of optic disc and optic cup using Fuzzy C Means clustering technique combined with
thresholding. Using the extracted optic disc and optic cup the vertical cup-to-disc ratio was calculated.
The validity of this new method has been tested on 365 colour fundus images from two different publicly
available databases DRION, DIARATDB0 and images from an ophthalmologist. The result of the method
seems to be promising and useful for clinical work.
The legal cause of blindness for the workingage
population in western countries is Diabetic Retinopathy - a
complication of diabetes mellitus - is a severe and wide- spread
eye disease. Digital color fundus images are becoming
increasingly important for the diagnosis of Diabetic Retinopathy.
In order to facilitate and improve diagnosis in different ways, this
fact opens the possibility of applying image processing techniques
.Microaneurysms is the earliest sign of DR, therefore an
algorithm able to automatically detect the microaneurysms in
fundus image captured. Since microaneurysms is a necessary
preprocessing step for a correct diagnosis. Some methods that
address this problem can be found in the literature but they have
some drawbacks like accuracy or speed. The aim of this thesis is
to develop and test a new method for detecting the
microaneurysms in retina images. To do so preprocessing, gray
level 2D feature based vessel extraction is done using neural
network by using extra neurons which is evaluated on DRIVE
database which is superior than rulebased methods. To identify
microaneurysms in an image morphological opening and image
enhancement is performed. The complete algorithm is developed
by using a MATLAB implementation and the diagnosis in an
image can be estimated with the better accuracy and in shorter
time than previous techniques
Automatic Detection of Diabetic Maculopathy from Funduas Images Using Image A...Eman Al-dhaher
Diabetic retinopathy is a severe eye disease that affects many diabetic patients. It changes the small blood vessels in the retina resulting in loss of vision. Early detection and diagnosis have been identified as one of the ways to achieve a reduction in the percentage of visual impairment and blindness caused by diabetic retinopathy with emphasis on regular screening for detection and monitoring of this disease.
The work focuses on developing a fundus image analysis system that extracts the fundal features of the retina such as optic disk, macula (i.e., fovea) and exudates lesions (hard and soft exudates), which are the fundamental steps in an automated analyzing system to display and diagnosis diabetic retinopathy.
The main cause of eye diseases in the working human is Diabetic retinopathy. Eye disease can
be prevented if detects early. The extraction of blood vessels from retinal images is an essential and challenging
task in medical diagnosis and analysis. This paper describes the effective and efficient extraction of blood
vessels from retinal image by using Kirsch’s templates. The Kirsch’s edge operators detect the edges using eight
filters, generated by the compass rotation mechanism. The method is used to automatic detection of landmark
features of the fundus, such as the optic disc, fovea and blood vessels.
Retinal image analysis using morphological process and clustering techniquesipij
This paper proposes a method for the Retinal image analysis through efficient detection of exudates and
recognizes the retina to be normal or abnormal. The contrast image is enhanced by curvelet transform.
Hence, morphology operators are applied to the enhanced image in order to find the retinal image ridges.
A simple thresholding method along with opening and closing operation indicates the remained ridges
belonging to vessels. The clustering method is used for effective detection of exudates of eye. Experimental
result proves that the blood vessels and exudates can be effectively detected by applying this method on the
retinal images. Fundus images of the retina were collected from a reputed eye clinic and 110 images were
trained and tested in order to extract the exudates and blood vessels. In this system we use the Probabilistic
Neural Network (PNN) for training and testing the pre-processed images. The results showed the retina is
normal or abnormal thereby analyzing the retinal image efficiently. There is 98% accuracy in the detection
of the exudates in the retina .
Teamed with 2 students to research and implement the automation of diagnosis of Diabetic Retinopathy and co-ordinated with an Ophthalmologist to verify our implementation.
Responsibilities included MATLAB coding, algorithm testing, and product documentation.
• Automation in MATLAB involving retinal image analysis to help
Ophthalmologist increase the productivity and efficiency in a clinical
environment.
• Used Image Processing concepts such as Hough Transform, Bottom Hat
Transform, Edge Detection Technique and Morphological Operators.
Provided our algorithm and documentation to our research faculty advisor to enable him to continue this research to the next phase.
IMAGE PROCESSING Projects for M. Tech, IMAGE PROCESSING Projects in Vijayanagar, IMAGE PROCESSING Projects in Bangalore, M. Tech Projects in Vijayanagar, M. Tech Projects in Bangalore, IMAGE PROCESSING IEEE projects in Bangalore, IEEE 2015 IMAGE PROCESSING Projects, MATLAB Image Processing Projects, MATLAB Image Processing Projects in Bangalore, MATLAB Image Processing Projects in Vijayangar
Segmentation of the Blood Vessel and Optic Disc in Retinal Images Using EM Al...iosrjce
IOSR Journal of Computer Engineering (IOSR-JCE) is a double blind peer reviewed International Journal that provides rapid publication (within a month) of articles in all areas of computer engineering and its applications. The journal welcomes publications of high quality papers on theoretical developments and practical applications in computer technology. Original research papers, state-of-the-art reviews, and high quality technical notes are invited for publications.
C LASSIFICATION O F D IABETES R ETINA I MAGES U SING B LOOD V ESSEL A REASIJCI JOURNAL
Retina images are obtained from the fundus camera a
nd graded by skilled professionals. However there i
s
considerable shortage of expert observers has encou
raged computer assisted monitoring. Evaluation of
blood vessels network plays an important task in a
variety of medical diagnosis. Manifestations of
numerous vascular disorders, such as diabetic retin
opathy, depend on detection of the blood vessels
network. In this work the fundus RGB image is used
for obtaining the traces of blood vessels and areas
of
blood vessels are used for detection of Diabetic Re
tinopathy (DR). The algorithm developed uses
morphological operation to extract blood vessels. M
ainly two steps are used: firstly enhancement opera
tion
is applied to original retina image to remove noise
and increase contrast of retinal blood vessels. Se
condly
morphology operations are used to take out blood ve
ssels. Experiments are conducted on publicly availa
ble
DIARETDB1 database. Experimental results obtained b
y using gray-scale images have been presented.
Automated histopathological image analysis: a review on ROI extractioniosrjce
IOSR Journal of Computer Engineering (IOSR-JCE) is a double blind peer reviewed International Journal that provides rapid publication (within a month) of articles in all areas of computer engineering and its applications. The journal welcomes publications of high quality papers on theoretical developments and practical applications in computer technology. Original research papers, state-of-the-art reviews, and high quality technical notes are invited for publications.
Review of methods for diabetic retinopathy detection and severity classificationeSAT Journals
Abstract Diabetic Retinopathy is a serious vascular disorder that might lead to complete blindness. Therefore, the early detection and the treatment are necessary to prevent major vision loss. Though the Manual screening methods are available, they are time consuming and inefficient on a large image database of patients. Moreover, it demands skilled professionals for the diagnosis. Automatic Diabetic Retinopathy diagnosis systems can replace manual methods as they can significantly reduce the manual labor involved in the screening process. Screening conducted over a larger population can become efficient if the system can separate normal and abnormal cases, instead of the manual examination of all images. Therefore, Automatic Retinopathy detection systems have attracted large popularity in the recent times. Automatic retinopathy detection systems employ image processing and computer vision techniques to detect different anomalies associated with retinopathy. This paper reviews various methods of diabetic retinopathy detection and classification into different stages based on severity levels and also, various image databases used for the research purpose are discussed. Keywords— Automatic Diabetic Retinopathy detection, computer vision, Diabetic Retinopathy, image databases, image processing, manual screening
Automatic detection of optic disc and blood vessels from retinal images using...eSAT Publishing House
IJRET : International Journal of Research in Engineering and Technology is an international peer reviewed, online journal published by eSAT Publishing House for the enhancement of research in various disciplines of Engineering and Technology. The aim and scope of the journal is to provide an academic medium and an important reference for the advancement and dissemination of research results that support high-level learning, teaching and research in the fields of Engineering and Technology. We bring together Scientists, Academician, Field Engineers, Scholars and Students of related fields of Engineering and Technology
FUZZY CLUSTERING BASED GLAUCOMA DETECTION USING THE CDR sipij
Glaucoma is a serious eye disease, overtime it will result in gradual blindness. Early detection of thedisease will help prevent against developing a more serious condition. A vertical cup-to-disc ratio which isthe ratio of the vertical diameter of the optic cup to that of the optic disc, of the fundus eye image is an important clinical indicator for glaucoma diagnosis. This paper presents an automated method for the extraction of optic disc and optic cup using Fuzzy C Means clustering technique combined with
thresholding. Using the extracted optic disc and optic cup the vertical cup-to-disc ratio was calculated.
The validity of this new method has been tested on 365 colour fundus images from two different publicly
available databases DRION, DIARATDB0 and images from an ophthalmologist. The result of the method
seems to be promising and useful for clinical work.
The legal cause of blindness for the workingage
population in western countries is Diabetic Retinopathy - a
complication of diabetes mellitus - is a severe and wide- spread
eye disease. Digital color fundus images are becoming
increasingly important for the diagnosis of Diabetic Retinopathy.
In order to facilitate and improve diagnosis in different ways, this
fact opens the possibility of applying image processing techniques
.Microaneurysms is the earliest sign of DR, therefore an
algorithm able to automatically detect the microaneurysms in
fundus image captured. Since microaneurysms is a necessary
preprocessing step for a correct diagnosis. Some methods that
address this problem can be found in the literature but they have
some drawbacks like accuracy or speed. The aim of this thesis is
to develop and test a new method for detecting the
microaneurysms in retina images. To do so preprocessing, gray
level 2D feature based vessel extraction is done using neural
network by using extra neurons which is evaluated on DRIVE
database which is superior than rulebased methods. To identify
microaneurysms in an image morphological opening and image
enhancement is performed. The complete algorithm is developed
by using a MATLAB implementation and the diagnosis in an
image can be estimated with the better accuracy and in shorter
time than previous techniques
Automatic Detection of Diabetic Maculopathy from Funduas Images Using Image A...Eman Al-dhaher
Diabetic retinopathy is a severe eye disease that affects many diabetic patients. It changes the small blood vessels in the retina resulting in loss of vision. Early detection and diagnosis have been identified as one of the ways to achieve a reduction in the percentage of visual impairment and blindness caused by diabetic retinopathy with emphasis on regular screening for detection and monitoring of this disease.
The work focuses on developing a fundus image analysis system that extracts the fundal features of the retina such as optic disk, macula (i.e., fovea) and exudates lesions (hard and soft exudates), which are the fundamental steps in an automated analyzing system to display and diagnosis diabetic retinopathy.
Automated Detection of Optic Disc in Retinal FundusImages Using PCAiosrjce
IOSR Journal of Pharmacy and Biological Sciences(IOSR-JPBS) is a double blind peer reviewed International Journal that provides rapid publication (within a month) of articles in all areas of Pharmacy and Biological Science. The journal welcomes publications of high quality papers on theoretical developments and practical applications in Pharmacy and Biological Science. Original research papers, state-of-the-art reviews, and high quality technical notes are invited for publications.
Abstract:—The main cause of eye diseases in the working human is Diabetic retinopathy. Eye disease can
be prevented if detects early. The extraction of blood vessels from retinal images is an essential and challenging
task in medical diagnosis and analysis. This paper describes the effective and efficient extraction of blood
vessels from retinal image by using Kirsch’s templates. The Kirsch’s edge operators detect the edges using eight
filters, generated by the compass rotation mechanism. The method is used to automatic detection of landmark
features of the fundus, such as the optic disc, fovea and blood vessels.
Keywords: —Diabetic retinopathy, Retinal image, Oculist
PSO-HRVSO: Segmentation of Retinal Vessels Through Homomorphic Filtering Enha...sipij
The structure of retinal blood vessels is crucial for the early detection of diabetic retinopathy, a leading
cause of blindness worldwide. Yet, accurately segmenting retinal vessels poses significant challenges due
to the low contrast and noise present in capillaries.The automated segmentation of retinal blood vessels
significantly enhances Computer-Aided Diagnosis for diverse ophthalmic and cardiovascular conditions. It
is imperative to develop a method capable of segmenting both thin and thick retinal vessels to facilitate
medical analysis and disease diagnosis effectively. This article introduces a novel methodology for robust
vessel segmentation, addressing prevalent challenges identified in existing literature.
PSO-HRVSO: SEGMENTATION OF RETINAL VESSELS THROUGH HOMOMORPHIC FILTERING ENHA...sipij
The structure of retinal blood vessels is crucial for the early detection of diabetic retinopathy, a leading
cause of blindness worldwide. Yet, accurately segmenting retinal vessels poses significant challenges due
to the low contrast and noise present in capillaries.The automated segmentation of retinal blood vessels
significantly enhances Computer-Aided Diagnosis for diverse ophthalmic and cardiovascular conditions. It
is imperative to develop a method capable of segmenting both thin and thick retinal vessels to facilitate
medical analysis and disease diagnosis effectively. This article introduces a novel methodology for robust
vessel segmentation, addressing prevalent challenges identified in existing literature.
The methodology PSO-HRVSO comprises three key stages: pre-processing, main processing, and postprocessing. In the initial stage, filters are employed for image smoothing and enhancement, leveraging
PSO optimization. The main processing phase is bifurcated into two configurations. Initially, thick vessels
are segmented utilizing an optimized top-hat approach, homo-morphic filtering, and median filter. Subsequently, the second configuration targets thin vessel segmentation, employing the optimized top-hat method, homomorphic filtering, and matched filter. Lastly, morphological image operations are conducted during the post-processing stage.
The PSO-HRVSO method underwent evaluation using two publicly accessible databases (DRIVE and
STARE), measuring performance across three key metrics: specificity, sensitivity, and accuracy. Analysis
of the outcomes revealed averages of 0.9891, 0.8577, and 0.0.9852 for the DRIVE dataset, and 0.9868,
0.8576, and 0.9831 for the STARE dataset, respectively.
The PSO-HRVSO technique yields numerical results that demonstrate competitive average values when
compared to current methods. Moreover, it sur-passes all leading unsupervised methods in terms of specificity and accuracy. Additionally, it outperforms the majority of state-of-the-art supervised methods without
incurring the computational costs associated with such algorithms. Detailed visual analysis reveals that
the PSO-HRVSO approach enables a more precise segmentation of thin vessels compared to alternative
procedures.
Detection of Glaucoma using Optic Disk and Incremental Cup Segmentation from ...theijes
Medical researchers, detection of eye disease is very important because it may causes blindness. Glaucoma is one of the diseases that cause blindness. Standard procedure for detection glaucoma is to analysis of optic disk (OD) and cup region in retinal image. In this paper, introduce an automatic OD parameterized technique which is based on segmentation and Incremental Cup segmentation. The incremental cup segmentation method is based on anatomical evidence such as vessel bends at the cup boundary, considered relevant by glaucoma experts. Bends in a vessel are robustly detected using a region of support concept, which automatically selects the right scale for analysis. A multi-stage strategy is applied to derive a reliable subset of vessel bends called r-bends followed by a local 2-D spline fitting to derive the desired cup boundary. The results are compared with existing methods using different retinal images.
AUTOMATED SEGMENTATION OF FLUORESCENT AND FUNDS IMAGES BASED ON RETINAL BLOOD...acijjournal
ABSTRACT
Measurements of retinal blood vessel morphology have been shown to be related to the risk of cardiovascular diseases. The wrong identification of vessels may result in a large variation of these measurements, leading to a wrong clinical diagnosis. In this paper, we address the problem of automatically identifying true vessels as a post processing step to vascular structure segmentation. We model the segmented vascular structure as a vessel segment graph and formulate the problem of identifying vessels as one of finding the optimal forest in the graph given a set of constraints. We design a method to solve this optimization problem and evaluate it on a large real-world dataset of 2,446 retinal images. Experiment results are analyzed with respect to actual measurements of vessel morphology. The results show that the proposed approach is able to achieve 98.9% pixel precision and 98.7% recall of the true vessels for clean segmented retinal images, and remains robust even when the segmented image is noisy.
AUTOMATED SEGMENTATION OF FLUORESCENT AND FUNDS IMAGES BASED ON RETINAL BLOOD...acijjournal
Measurements of retinal blood vessel morphology have been shown to be related to the risk of
cardiovascular diseases. The wrong identification of vessels may result in a large variation of these
measurements, leading to a wrong clinical diagnosis. In this paper, we address the problem of
automatically identifying true vessels as a post processing step to vascular structure segmentation. We
model the segmented vascular structure as a vessel segment graph and formulate the problem of identifying
vessels as one of finding the optimal forest in the graph given a set of constraints. We design a method to
solve this optimization problem and evaluate it on a large real-world dataset of 2,446 retinal images.
Experiment results are analyzed with respect to actual measurements of vessel morphology. The results
show that the proposed approach is able to achieve 98.9% pixel precision and 98.7% recall of the true
vessels for clean segmented retinal images, and remains robust even when the segmented image is noisy.
The International Journal of Engineering & Science is aimed at providing a platform for researchers, engineers, scientists, or educators to publish their original research results, to exchange new ideas, to disseminate information in innovative designs, engineering experiences and technological skills. It is also the Journal's objective to promote engineering and technology education. All papers submitted to the Journal will be blind peer-reviewed. Only original articles will be published.
Retina is a layer which is found at the back side of the eye ball which plays main role for visualization. Any
disease in the retina leads to severe problems. Blood vessels segmentation and classification of retinal
vessels into arteries and veins is an essential thing for detection of various diseases like Diabetic
Retinography etc. This paper discusses about various existing methodologies for classification of retinal
image into artery and vein which are helpful for the detection of various diseases in retinal fundus image.
This process is basis for the AVR calculation i.e. for the calculation of average diameter of arteries to
veins. One of the symptoms of Diabetic Retinography causes abnormally wide veins and this leads to low
ratio of AVR. Diseases like high blood pressure and pancreas also have abnormal AVR. Thus classification
of blood vessels into arteries and veins is more important. Retinal fundus images are available on the
publically available Database like DRIVE [5], INSPIREAVR [6], VICAVR [7].
Forklift Classes Overview by Intella PartsIntella Parts
Discover the different forklift classes and their specific applications. Learn how to choose the right forklift for your needs to ensure safety, efficiency, and compliance in your operations.
For more technical information, visit our website https://intellaparts.com
Water scarcity is the lack of fresh water resources to meet the standard water demand. There are two type of water scarcity. One is physical. The other is economic water scarcity.
TECHNICAL TRAINING MANUAL GENERAL FAMILIARIZATION COURSEDuvanRamosGarzon1
AIRCRAFT GENERAL
The Single Aisle is the most advanced family aircraft in service today, with fly-by-wire flight controls.
The A318, A319, A320 and A321 are twin-engine subsonic medium range aircraft.
The family offers a choice of engines
COLLEGE BUS MANAGEMENT SYSTEM PROJECT REPORT.pdfKamal Acharya
The College Bus Management system is completely developed by Visual Basic .NET Version. The application is connect with most secured database language MS SQL Server. The application is develop by using best combination of front-end and back-end languages. The application is totally design like flat user interface. This flat user interface is more attractive user interface in 2017. The application is gives more important to the system functionality. The application is to manage the student’s details, driver’s details, bus details, bus route details, bus fees details and more. The application has only one unit for admin. The admin can manage the entire application. The admin can login into the application by using username and password of the admin. The application is develop for big and small colleges. It is more user friendly for non-computer person. Even they can easily learn how to manage the application within hours. The application is more secure by the admin. The system will give an effective output for the VB.Net and SQL Server given as input to the system. The compiled java program given as input to the system, after scanning the program will generate different reports. The application generates the report for users. The admin can view and download the report of the data. The application deliver the excel format reports. Because, excel formatted reports is very easy to understand the income and expense of the college bus. This application is mainly develop for windows operating system users. In 2017, 73% of people enterprises are using windows operating system. So the application will easily install for all the windows operating system users. The application-developed size is very low. The application consumes very low space in disk. Therefore, the user can allocate very minimum local disk space for this application.
Quality defects in TMT Bars, Possible causes and Potential Solutions.PrashantGoswami42
Maintaining high-quality standards in the production of TMT bars is crucial for ensuring structural integrity in construction. Addressing common defects through careful monitoring, standardized processes, and advanced technology can significantly improve the quality of TMT bars. Continuous training and adherence to quality control measures will also play a pivotal role in minimizing these defects.
Sachpazis:Terzaghi Bearing Capacity Estimation in simple terms with Calculati...Dr.Costas Sachpazis
Terzaghi's soil bearing capacity theory, developed by Karl Terzaghi, is a fundamental principle in geotechnical engineering used to determine the bearing capacity of shallow foundations. This theory provides a method to calculate the ultimate bearing capacity of soil, which is the maximum load per unit area that the soil can support without undergoing shear failure. The Calculation HTML Code included.
CFD Simulation of By-pass Flow in a HRSG module by R&R Consult.pptxR&R Consult
CFD analysis is incredibly effective at solving mysteries and improving the performance of complex systems!
Here's a great example: At a large natural gas-fired power plant, where they use waste heat to generate steam and energy, they were puzzled that their boiler wasn't producing as much steam as expected.
R&R and Tetra Engineering Group Inc. were asked to solve the issue with reduced steam production.
An inspection had shown that a significant amount of hot flue gas was bypassing the boiler tubes, where the heat was supposed to be transferred.
R&R Consult conducted a CFD analysis, which revealed that 6.3% of the flue gas was bypassing the boiler tubes without transferring heat. The analysis also showed that the flue gas was instead being directed along the sides of the boiler and between the modules that were supposed to capture the heat. This was the cause of the reduced performance.
Based on our results, Tetra Engineering installed covering plates to reduce the bypass flow. This improved the boiler's performance and increased electricity production.
It is always satisfying when we can help solve complex challenges like this. Do your systems also need a check-up or optimization? Give us a call!
Work done in cooperation with James Malloy and David Moelling from Tetra Engineering.
More examples of our work https://www.r-r-consult.dk/en/cases-en/
Segmentation of Blood Vessels and Optic Disc in Retinal Images
1. Research Inventy: International Journal of Engineering And Science
Vol.6, Issue 5 (May 2016), PP -34-42
Issn (e): 2278-4721, Issn (p):2319-6483, www.researchinventy.com
34
Segmentation of Blood Vessels and Optic Disc in Retinal Images
Kota Prajwal Kant
Gandhi institute of advanced computer and research Dept of ECE,
Abstract: Retinal image analysis is increasingly prominent as a non-intrusive diagnosis method in
modern ophthalmology. In this paper, we present a novel method to segment blood vessels and optic
disc in the fundus retinal images. The method could be used to support non-intrusive diagnosis in
modern ophthalmology since the morphology of the blood vessel and the optic disc is an important
indicator for diseases like diabetic retinopathy, glaucoma and hypertension. Our method takes as
first step the extraction of the retina vascular tree using the graph cut technique. The blood vessel
information is then used to estimate the location of the optic disc. The optic disc segmentation is
performed using two alternative methods. The Markov Random Field (MRF) image reconstruction
method segments the optic disc by removing vessels from the optic disc region and the
Compensation Factor method segments the optic disc using prior local intensity knowledge of the
vessels. The proposed method is tested on three public data sets, DIARETDB1, DRIVE and STARE.
The results and comparison with alternative methods show that our method achieved exceptional
performance in segmenting the blood vessel and optic disc.
Index Terms: Retinal images, vessel segmentation, optic disc segmentation, graph cut segmentation.
I. Introduction
The segmentation of retinal image structure has been of great interest because it could as a
non intrusive diagnosis in modern ophthalmology. The morphology of the retinal blood vessel and the
optic disc is an important structural indicator for assessing the presence and severity of retinal
diseases such as diabetic retinopathy, hypertension, glaucoma, haemorrhages, vein occlusion and
neo-vascularisation. However to assess the diameter and tortuosity of retinal blood vessel or the
shape of the optic disc, manual planimetry has commonly been used by ophthalmologist, which is
generally time consuming and prone with human error, especially when the vessel structure are
complicated or a large number of image are acquired and prone with human error , especially when
the vessel structure are compited or a large number of images are acquired to be labeled by hand .
therefore a reliable automated methos for retinal blood vessel and optic disc segmentation, which
preserves various vessel and optic disc characteristics is attractive in computer aided-diagnosis.
An automated segmentation and inspection of retinal blood vessel features such as diameter, colour
and tortuosity as well as the optic disc morphology allows ophthalmologist and eye care specialists
to perform mass vision screening exams for early detection of retinal diseases and treatment
evaluation. This could prevent and reduce vision impairments; age related diseases and many
cardiovascular diseases as well as reducing the cost of the screening.
Over the past few years, several segmentation techniques have been employed for the segmentation of
retinal structures such as blood vessels and optic disc and diseases like lesions in fundus retinal images. However
the acquisition of fundus retinal images under different conditions of illumination, resolution and field of view
(FOV) and the overlapping in the retina cause a significant to the performance of automated blood vessel and
optic disc segmentations. Thus, there is a need for a reliable technique for retinal vascular tree extraction and
optic disc detection, which preserves various vessel and optic disc shapes. In the following segment, we briefly
review the previous studies on blood vessel segmentation and optic disc segmentation separately.
II. Related Works
Two different approaches have been developed to segment the vessels of the retina. The pixel
processing based method and tracking method The pixel processing based approach performs the vessel
segmentation in a two-pass operation. First the appearance of the vessel is enhanced using detection process
such as morphological pre processing techniques and adaptive filtering. The second operation is the recognition
of vessel structure Using thinning or branch point operation to classify a pixel as a vessel background. These
approaches process every pixel in the image apply multiple operation on each pixel. The second set of
approaches to vessel segmentation are referred to as vessel tracking, vectorial tracking or tracing [1]. In contrast
to the pixel processing based approaches, the tracking methods detect first initial vessel seed points, and then
track the rest of the vessel pixels through the image by measuring the continuity proprieties of the blood
vessels. This technique is used as a single pass operation, where the detection of the vessel structures and the
recognition of the structures are simultaneously performed.
2. Segmentation of Blood Vessels and Optic Disc in Retinal Images
35
The tracking based approaches included semi automated tracing and automated tracing. In the semi
automated tracing methods, the user manually selects the initial vessel seed point. These methods are generally
used in quantitative coronary angiography analysis and they generally provide accurate segmentation of the
vessels. In fully automated tracing, the algorithms automatically select the initial vessel points and most
methods use Gaussian functions to characterise a vessel profile model, which locates a vessel points for the
vessel tracing. They are computationally efficient and more suitable for retinal image processing. Examples of
the tracking based approaches are presented in Xu et al. [8], Maritiner-perez et al. [9], Staal et al. [5], Zhou et al
[10].
Both pixel processing and tracking approaches have their own advantages and limitations over each
other. The pixel processing approaches can provide a complete extraction of the vascular tree in the retinal
image since they search all the possible vessel pixels across the whole image. However these techniques are
computationally expensive and require special hardware to be suitable for large image dataset. The presence of
noise and lesions in some retinal images causes a significant degradation in the performance of the pixel
processing approaches as the enhancement operation may pick up some noise and lesions as vessel pixels. This
could lead to false vessel detection in the recognition operation. On the other hand, the tracking approaches are
computationally efficient and much faster than the pixels processing methods because they perform the vessel
segmentation using only the pixels in the neighbourhood of the vessels structure and avoid the processing of
every pixel in the image. Nevertheless, these methods lack in extracting a complete vascular tree in the case
where there are discontinuities in the vessel branches. Further more, the semi automated tracking segmentation
methods need manual input, which requires time.
The optic nerve head is described as the brightest round area in the retina where the blood vessels
converge with a shape that is approximately elliptical and has a width of 1.8 ± 0.2 mm and height 1.9 ± 0.2
mm [11]. The convergence feature of blood vessels into the optic disc region is generally used to estimate the
location of the optic disc and segment it from the retinal image. But the intrusion of vessels in the optic disc
region constitutes computational complexity for the optic disc segmentation as it is breaking the continuity of
its boundary. To address this problem, several methods have been employed such as Chrastek et al. [12],
Lowell et al. [13], Welfer et al. [14] and Aquino et al. [15] . Chrastek et al. [12] presented an automated
segmentation of the optic nerve head for diagnosis of glaucoma. The method removes the blood vessel by using
a distance map.
Algorithm, then the optic disc is segmented by combining a morphological operation, Hough
Transform and an anchored active contour model. Lowell et al. [13] proposed a deformable contour model to
segment the optic nerve head boundary in low resolution retinal images. The approach localises the optic
disc using a specialised template matching and a directionally-sensitive gradient to eliminate the obstruction
of the vessel in the optic disc region before performing the segmentation. Welfer et al. [14] proposed an
automated segmentation of the optic disc in colour eye fundus image using an adaptive morphological
operation. The method uses a watershed transform markers to define the optic disc boundary and the vessel
obstruction is minimised by morphological erosion.
These techniques are performed using morphological operations to eliminate the blood vessels from
the retinal image. However, the application of morphological operations can modify the image by corrupting
some useful information.
In our optic disc segmentation process, the convergence feature of vessels into the optic disc region
is used to estimate its location. We then use two automated methods (Markov Random field image
reconstruction and Compensation Factor) to segment the optic disc.
The rest of the paper is organised as follow. The blood vessel segmentation is discussed in Section III.
Section IV provides the detailed description of the optic disc segmenta- tion. Section V presents the
experimental results of our method with comparisons to other methods. Conclusions are drawn in Section VI.
The preliminary results of the three components of the approach, namely the blood vessel segmentation, optic
disc segmentation using the Graph Cut and Markov Random Field respectively, were presented separately in
[16], [17], [18]. More details of the approach can be found in the PhD thesis [19].
III. Blood Vessels Segmentation
Blood vessels can be seen as thin elongated structures in the retina, with variation in width and length.
In order to segment the blood vessel from the fundus retinal image, we have implemented a pre-processing
technique, which consists of effective adaptive histogram equalisation (AHE) and robust distance transform.
This operation improves the robustness and the accuracy of the graph cut algorithm. Fig. 1 shows the
illustration of the vessel segmentation algorithm.
3. Segmentation of Blood Vessels and Optic Disc in Retinal Images
36
Fig. 1. Vessel segmentation algorithm
A. Pre-processing
We apply a contrast enhancement process to the green channel image similar to the work presented in [20]. The intensity of
the image is inverted, and the illumination is equalised. The resulting image is enhanced using an adaptive histogram
equaliser, given by:
EQ1
where I is the green channel of the fundus retinal colour image, p denotes a pixel and p0
is the neighbourhood
pixel around p. p0
2 R(p) is the square window neighbourhood with length h. s(d) = 1 if d > 0, and s(d) = 0
otherwise with d = s (I (p) I (p0
)). M = 255 value of the maximum intensity in the image. r is a parameter to
control the level of enhancement. Increasing the value of r would also increase the contrast between vessel
pixels and the background (see Fig. 2). The experimental values of the window length was set to h = 81 and r =
6. A binary morphological open process is applied to prune the enhanced image, which discards all the
misclassified pixels see (Fig. 2 (d)). This approach significantly reduces the false positive, since the enhanced
image will be used to construct the graph for segmentation.
A distance map image is created using the distance transform algorithm. This is used to calculate the direction
and magnitude of the vessel gradient. Fig. 2 (e) and (f) show the distance map of the whole image and a sample
vessel with arrows indicating the direction of the gradients respectively. From the sample vessel image, we can
see the centre line with the brightest pixels, which are progressively reduced in intensity in the direction of the
edges (image gradients). The arrows in Fig. 2 (f) referred as vector field, which is used to construct the graph in
the next Sections.
A. Graph construction for vessel segmentation
The graph cut is an energy based object segmentation approach. The technique is characterised by
an optimization operation designed to minimise the energy generated from a given image data. This energy
defines the relationship between neighborhood pixel elements in an image.
A graph G ( ; ) is defined as a set of nodes (pixels) and a set of undirected edges that connect these
neighbouring nodes. The graph included two special nodes, a foreground terminal (source S) and a background
terminal (sink T). includes two types of undirected edges: neighbourhood links (n-links) and terminal links (t-
links). Each pixel p 2 P (a set of pixels) in the graph presents two t-links fp; Sg and fp; T g connecting it to each
terminal while a pair of neighbouring pixels fp; qg 2 N (number of pixel neighbour) is connect by a n-links [21].
Thus:
EQ 2
4. Segmentation of Blood Vessels and Optic Disc in Retinal Images
37
An edge e 2 is assigned a weight (cost) We> 0. A cut is defined by a subset of edges C 2 where G (c) = h ; nCi
separating the graph into two foreground and background with
EQ 3
The graph cut technique is used in our segmentation because it allows the incorporation of prior knowledge into
the graph formulation in order to guide the model and find the optimal segmentation. Let assume A = (A1; Ap; : :
: AP ) a binary vector set of labels assigned to each pixel p in the image, where Ap indicate assignments to pixels
p in P . Therefore, each assignment Ap is either in foreground (F g) or background (Bg). Thus the segmentation
is obtained by the binary vector A and the constraints imposed on the regional and boundary proprieties of
vector A are derived by the energy formulation of the graph defined as
EQ 4
where the positive coefficient λ indicates the relative impor- tance of the regional term (likelihoods of
foreground and back- ground) RA against the boundary term (relationship between neighbourhood pixels) BA.
The regional or the likelihood of the foreground and background is given by
EQ 5 6 7
During the minimisation of the graph energy formulation in
to segment thin objects like blood vessels, the second term (boundary term) in (4) has a tendency to follow short
edges known as “the shrinking bias” [22]. This problem causes a significant degradation on the performance of
the graph cut algorithm on thin elongated structures like the blood vessels. Fig. 3 shows an example of the blood
vessel segmentation using the traditional graph formulation [23]. From Fig. 3, it can be seen that the blood
vessel segmentation follows short edges, and tends to shrink in the searching for the cheapest cost. It can also be
noticed that λ in (4) controls the relation between boundary and regional terms. Increasing the value of λ, the
likelihood of the pixels belonging to foreground and
background (t-links) gains strength over the regional term (n-links), which slightly improved the
segmentation result see Fig. 3 (d).
To address the above problem, the segmentation of blood vessels using the graph cut requires special graph formulation. One
ofthe method used to address the shrinking bias problem
described geometric proprieties of the discrete cut metric on regular grids and Finsler length can be represented
by the sum of two terms. Those terms represent the symmetric and anti-symmetric parts of the cut metric. The
symmetric part of the cut defines the standard geometric length of contour and it is independent of its
orientation. The anti-symmetric part of the cut metric represents the flux of a given vector field through the
contour [23].
To address “the shrinking bias” problem seen in Fig. 3, we have constructed a graph consisting of a symmetric part
g+ (shrinking) and an anti-symmetric part g− (stretching) by incorporating the flux of vector v into the graph
construction. The symmetric part g+ of the graph corresponds to a cut geometric length and is related directly
5. Segmentation of Blood Vessels and Optic Disc in Retinal Images
38
k
k
with the n-link connections and the anti-symmetric part g− is equal to flux of vector field v over the cut
geometric and it is used to derive the t-links. Thus the the blood vessels can be segment by keeping a good
balance between shrinking and stretching (flux) throughout the image boundary.
1) The symmetric part of the graph: is used to assign weights on the n-link connections (edges
2) between neigh- bouring pixels). Let consider a neighbour system of a graph described by a set of edges ek ,
where 1 ≤ k ≤ N , for N number of neighbours. Let us define ek as the shortest vector
connecting two pixels in the direction of k, W +(p) the weight
of the edge ek at pixel p and Wf+(p) a set of the edge weights at pixel p for all directions. The corresponding
edge weights are defined by
3) The anti-symmetric part of the graph : We used the term Anti-Symmetry because, the flux (stretching) of
vector field v over the cut geometric balanced the shrinking of blood vessels during the segmentation. This
anti-symmetric part of the graph is defined by the flux of vector field v over the cut geometric. It is used to
assign weights on the t-links (edges between a given pixel and the terminals) to balance the shrinking effect
seen in Fig. 3. Specific weights for t-links are obtained based on the deposition of vector v. Different
decompositions of vectorv may result in different t-links whose weights can be interpreted as an estimation
of divergence. In our implementation, we decomposed the vector v along grid edges with the n-links
oriented along the main axes, X and Y direction. Thus vector v can be decomposed as v = vxux + vyuy where
ux anduy are unit vectors in X
EQ 10
where vx
right
and vx
right
are the components of vector v in X direction taken at the right and left neighbour of pixel
P respectively. vy
up
and vy
down
are the Y of vector v taken at the top and down of of pixel P . is the size of the cell
in the grid map (see Fig. 5). We add edge (s ! p) with weight C ( tp) if tp < 0, or edge (p ! t) with weight C tp
otherwise. The parameter C is related to the magnitude of the vector v, thus pixels in the centre of the blood
vessel have a higher connection to the source (foreground) than pixels in the edge of the blood vessels. Because
the distance map is calculated on the pruned image and vector v is only defined for the pixels detected as blood
vessels in the rough segmentation. For the rest of the pixels in the image, the initialisation of t-link weights is set
as (p ! s) with weight t = 0 and (p ! t) with weight t = K, where K is the maximum weight sum for a pixel in the
symmetric construction. Fig. 6 shows the segmentation results of the blood vessels using different
decomposition of the vector v generating different t-link weights
Optic Disc Segmentation
The optic disc segmentation starts by defining the location of the optic disc. This process used the
convergence feature of vessels into the optic disc to estimate its location. The disc area is then segmented using
two different automated methods (Markov Random field image reconstruction and Compensation Factor). Both
methods use the convergence feature of the vessels to identify the position of the disc. The Markov Random
Field (MRF) method is applied to eliminate the vessel from the optic disc region. This process is known as
image reconstruction and it is performed only on the vessel pixels to avoid the modification of other structures
of the image. The reconstructed image is free of vessel and it is used to segment the optic disc via graph cut. In
contrast to MRF method, the Compensation Factor approach segments the optic disc using prior local intensity
knowledge of the vessels. Fig. 7 shows the overview of both the MRF and the Compensation Factor method
process.
6. Segmentation of Blood Vessels and Optic Disc in Retinal Images
39
FIG6
image of vessels segmented in Section III to find the location of the optic disc. The process iteratively trace
towards the centroid of the optic disc. The vessel image is pruned using a morphological open process to
eliminate thin vessels and keep the main arcade. The centroid of the arcade is calculated using the following
formulation:
K xi K yi
X Cy
=
Xi (12)
Cx
=
K K
i=1 =1
where xi and yi are the coordinates of the pixel in the binary image and K is the number of pixels set to 1 (pixels
marked as blood vessels) in the binary image.
Given the gray scale intensity of a retinal image, we select 1% of the brightest region. The algorithm
detects the brightest region with the most number of pixels to determine the location of the optic disc with
respect to the centroid point (right, left, up or down). The algorithm adjusts the centroid point iteratively until it
reaches the vessel convergence point or centre of the main arcade (centre of the optic disc) by reducing the
distance from one centroid point to next one in the direction of the brightest region, and correcting the central
position inside the arcade accordingly. Fig. 8 shows the process of estimating the location of the of optic disc in
a retinal image. It is important to notice that, the vessel convergence point must be detected accurately, since
this point is used to automatically mark foreground seeds. A point on the border of the optic disc may result in
some false foreground seeds. After the detection of the vessel convergence point, the image is constrained a
region of interest (ROI) including the whole area of the optic disc to minimize the processing time. This ROI is
set to a square of 200 by 200 pixels concentric with the detected optic disc centre. Then an automatic
initialisation of seeds (foreground and background) for the graph is performed. A neighbourhood of 20 pixels of
radius around the centre of the optic disc area is marked as foreground pixels and a band of pixels around the
perimeter of the image are selected as background seeds (see Fig. 9).
B. Optic Disc Segmentation with Markov Random Field Image Reconstruction
The high contrast of blood vessels inside the optic disc presented the main difficulty for it segmentation
as it misguides the segmentation through a short path, breaking the continuity of the optic disc boundary. To
address this problem, the MRF based reconstruction method presented in [25] is adapted in our work. We have
selected this approach because of its robustness. The objective of our algorithm is to find a best match for some
missing pixels in the image, however one of the weaknesses of MRF based reconstruction is the requirement of
intensive computation. To overcome this problem, we have limited the reconstruction to the region of interest
(ROI) and using prior segmented retina vascular tree, the reconstruction was performed in the ROI. An
overview diagram of the optic disc segmentation with Markov Random Field Image Reconstruction is shown in
Fig. 6.
FIG 7,8
Let us consider a pixel neighbourhood w(p) define as a square window of size W , where pixel p is the
centre of the neighbourhood . I is the image to be reconstructed and some of the pixels in I are missing. Our
objective is to find the best approximate values for the missing pixels in I. So let d(w1; w2) represent a
perceptual distance between two patches that defines their similarity. The exact matching patch corresponds to
d(w0
; w(p)) = 0. If we define a set of these patches as (p) = f!0
I : d(!0
; !(p)) = 0g the probability density function
of p can be estimated with a histogram of all centre pixel values in (p). However we are considering a finite
7. Segmentation of Blood Vessels and Optic Disc in Retinal Images
40
neighbourhood for p and the searching is limited to the image area, there might not be any exact matches for a
patch. For this reason, we find a collection of patches, which match falls between the best match and a
threshold. The closest match is calculated as !best = argmin!d(!(p); !) I. All the patches ! with d(!(p); !) < (1 +
)d(!(p); !best) are included in the collection !0
. d(w0
; w(p)) is defined as the sum of the absolute differences of the
intensities between patches, so identical patches will result in d(w0
; w(p)) = 0. Using the collection of patches,
we create an histogram and select the one with the highest mode. Fig. 10 shows sample results of the
reconstruction. The foreground F gs and the background
The graph cut algorithm descripted in section III-B is used to separate the foreground and the
background by minimising the energy function over the graph and producing the optimal segmentation of the
optic disc in the image. The energy function of the graph in (4) consists of regional and boundary terms. The
regional term (likelihoods of foreground and background) is calculated using (5), while the boundary term
(relationship between neighbouring pixels) is derived using (6). A grid of 16 neighbours N is selected to create
links between pixels in the image Im. The Max-Flow algorithm is used to cut the graph and find the optimal
segmentation.
C. Optic Disc Segmentation With Compensation Factor
In contrast to MRF image reconstruction, we have incorporated the blood vessels into the graph cut
formulation by introducing a compensation factor V ad. This factor is derived using prior information of blood
vessel.
The energy function of the graph cut algorithm generally comprises a boundary and regional terms. The
boundary term defined in (6) is used to assign weights on the edges (n-links) to measure the similarity between
neighbouring pixels with respect to the pixel proprieties (intensity, texture, colour). Therefore pixels with
similar intensities have a strong connec-tion. The regional term in (5) is derived to define the likelihood of the
pixel belonging to the background or the foreground by assigning weights on the edges (t-link) between image
pixels and the two terminals background and foreground seeds. In order to incorporated the blood vessels into
the graph cut formulation, we derived the t-link as follows:
ln Pr (IpnF gseeds) + V ad if p = vessel
S
link =
ln Pr (IpnF gseeds) if p 6= vessel
The intensity distribution of the blood vessel pixels in the region around the optic disc makes
them more likely to belong to background pixels than the foreground (or the optic disc pixels). Therefore the
vessels inside the disc have weak connections with neighbouring pixels making them likely to be segmented by
the graph cut as background. We introduce in (13) a compensation vector to all t-links of the foreground for
pixels belong to the vascular tree to address this behaviour. Consequently, vessels inside the optic disc are
classified with respect to their neighbourhood connections instead of their likelihood with the terminals
foreground and background seeds. Fig. 11 shows sample of images segmented by Compensation Factor. The
segmentation of the disc is affected by the value of V ad, the method achieves poor segmentation results for low
value of V ad. However when the value of the V ad increases, the performance improves until the value of V ad
is high enough to segment the rest of the vessels as foreground.
IV. Results
For the vessel segmentation method, we tested our algorithm on two public datasets, DRIVE [5],
STARE [2] with a total of 60 images. The optic disc segmentation algorithm was tested on DRIVE [5] and
DIARETDB1 [26], consisting of 129 images in total. The performance of both methods is tested against a number
of alternative methods.
The DRIVE consists of 40 digital images which were captured from a Canon CR5 non-mydriatic
3CCD camera at 45◦ field of view (FOV). The images have a size of 768 × 584
whereas the set B provides the manually labelled images for half of the dataset. To test our method
we adopt the set A hand labelling as the benchmark. We manually delimited the optic disc to test the
performance of optic disc segmentation algorithm.
The STARE dataset consists of 20 images captured by a TopCon TRV-50 fundus camera at 35◦ FOV.
The size of the images is 700 × 605 pixels. We calculated the mask image for this dataset using a simple
threshold technique for each
colour channel. The STARE dataset included images with retinal diseases selected by Hoover et al [2].
It also provides two sets of hand labelled images performed by two human experts. The first expert labelled
fewer vessel pixels than the second one. To test our method we adopt the first expert hand labelling as the
ground truth.
8. Segmentation of Blood Vessels and Optic Disc in Retinal Images
41
The DIARETDB1 dataset consist of 89 colour images with 84 of them contain at last one indication
of lesion. The images were captured with digital fundus camera at 50 degree filed of view and have a size of
1500 × 1152 pixels. Hand labelled lesion regions are provided in this dataset by four human experts. However
the DIARETDB1 dataset only
includes the hand labelled ground truth of lesions but not the blood vessels and the optic disc. For this
reason, we were unable to compare the performance of the blood vessel segmentation on the DIARETDB1
dataset. Nevertheless we were able to create the hand labelled ground truth of optic disc to test the performance
of the optic disc segmentation.
To facilitate the performance comparison between our method and alternative retinal blood vessels
segmentation approaches, parameters such as the true positive rate (TPR), the false positive rate (FPR) and the accuracy rate
(ACC) are derived to measure the performance of the segmentation [5]. The accuracy rate is defined as the sum of the
true positives (pixels correctly classified as vessel points) and the true negatives (non-vessel pixels correctly identified as
non vessel points), divided by the total number of pixel in the images. True Positive Rate (TPR) is defined as the total
number of true positives, divided by the number of blood vessel pixel marked in the ground true image. False Positive
Rate (FPR) is calculated as the total number of false positives divided by the number of pixels marked as non-vessel in the
ground true image. It is worth mentioning that a perfect segmentation would have a FPR of0 and a TPR of1. Our method and
all the alternative methods used the first expert hand labelled images as performance reference.
Most of the alternative methods use the whole image to measure the performance. In [5] all the
experiments are done on the FOV without considering the performance in the dark area outside the FOV. The
method in [3] measures the performance on both the whole image and the FOV. The
For the optic disc segmentation Tables V and VI present the performance of our method on
DIARETDB1 and DRIVE images. The results show that our methods of using (the Compensation factor and the
MRF image reconstruction) achieved the best overall performance. The results also show that, the MRF image
reconstruction algorithm outperforms the Compensation factor algorithm by 2.56% and 11.5% on DIARETDB1
and DRIVE images respectively. However it is important to notice that, the MRF image reconstruction
algorithm depends on the vessel segmentation algorithm, for example if the vessel segmentation algorithm
achieved a low performance on severely damage retinal image, the reconstruction would not define a
meaningful optic disc region, hence the segmentation will fail.
Further more, the proposed method addresses one of the main issues in medical image analysis, “the
overlapping tissue segmentation”. Since the blood vessels converse into the optic disc area and misguide the
graph cut algorithm through a short path, breaking the optic disc boundary. To achieve a good segmentation
results, the MRF image reconstruction algorithm eliminates vessels in the optic disc area without any
modification of the image structures before segmenting the optic disc. On the other hand the compensation
factor incorporates vessels using local intensity characteristic to perform the optic disc segmentation. Thus our
method can be applied in other medical image analysis applications to overcome “the overlapping tissue
segmentation.”
Our future research will be based on the segmentation of retinal diseases (lesions) known as
“exudates” using the segmented structures of the retina (blood vessels and optic disc).Thus a background
template can be created using these structures. Then this template can be used to perform the detection of
suspicious areas (lesions) in the retinal images.
References
[1]. K. Fritzsche, A. Can, H. Shen, C. Tsai, J. Turner, H. Tanenbuam, C. Stewart, B. Roysam, J. Suri, and S. Laxminarayan, “Automated
model based segmentation, tracing and analysis of retinal vasculature from digital fundus images,” State-of-The-Art Angiography,
Applications and Plaque Imaging Using MR, CT, Ultrasound and X-rays, pp. 225–298, 2003.
[2]. Hoover, V. Kouznetsova, and M. Goldbaum, “Locating blood vessels in retinal images by piecewise threshold probing of a matched
filter response,” IEEE Transactions on Medical Imaging, vol. 19, no. 3, pp. 203–210, 2000.
[3]. M. Mendonca and A. Campilho, “Segmentation of retinal blood vessels by combining the detection of centerlines and
morphological reconstruction,” IEEE Transactions on Medical Imaging, vol. 25, no. 9, 1200–1213, 2006.
[4]. J. Soares, J. Leandro, R. Cesar, H. Jelinek, and M. Cree, “Retinal vessel segmentation using the 2-d gabor wavelet and supervised
classification,”
[5]. IEEE Transactions on Medical Imaging, vol. 25, no. 9, pp. 1214–1222, 2006.
[6]. J. Staal, M. D. Abramoff, M. Niemeijer, M. A. Viergever, and B. van Ginneken, “Ridge-based vessel segmentation in color images
of the retina,” IEEE Transactions on Medical Imaging, vol. 23, no. 4, pp. 501– 509, 2004.
[7]. S. Chaudhuri, S. Chatterjee, N. Katz, M. Nelson, and M. Goldbaum, “Detection of blood vessels in retinal images using two-
dimensional matched filters,” IEEE Transactions on medical imaging, vol. 8, no. 3, 263–269, 1989.
[8]. F. Zana and J.-C. Klein, “Segmentation of vessel-like patterns using mathematical morphology and curvature evaluation,” Image
Processing, IEEE Transactions on, vol. 10, no. 7, pp. 1010–1019, 2001.
[9]. L. Xu and S. Luo, “A novel method for blood vessel detection from retinal images,” Biomedical engineering online, vol. 9, no. 1, p.
14, 2010.
[10]. M. E. Martinez-Perez, A. D. Hughes, S. A. Thom, A. A. Bharath, and K. H. Parker, “Segmentation of blood vessels from red-free
and fluorescein retinal images,” Medical image analysis, vol. 11, no. 1, pp. 47–61, 2007.
9. Segmentation of Blood Vessels and Optic Disc in Retinal Images
42
[11]. L. Zhou, M. S. Rzeszotarski, L. J. Singerman, and J. M. Chokreff, “The detection and quantification of retinopathy using digital
angiograms,”
[12]. Medical Imaging, IEEE Transactions on, vol. 13, no. 4, pp. 619–626, 1994.
[13]. Sinthanayothin, J. F. Boyce, H. L. Cook, and T. H. Williamson, “Automated localisation of the optic disc, fovea, and retinal blood
vessels from digital colour fundus images,” British Journal of Ophthalmology, vol. 83, no. 8, pp. 902–910, 1999.
[14]. R. Chrastek, M. Wolf, K. Donath, H. Niemann, D. Paulus, T. Hothorn, B. Lausen, R. Lammer, C. Y. Mardin, and G. Michelson,
“Automated segmentation of the optic nerve head for diagnosis of glaucoma,” Medical Image Analysis, vol. 9, no. 1, pp. 297–314,
2005.
[15]. J. Lowell, A. Hunter, D. Steel, A. Basu, R. Ryder, E. Fletcher, and L. Kennedy, “Optic nerve head segmentation,” IEEE
Transactions on Medical Imaging, vol. 23, no. 2, pp. 256–264, 2004.
[16]. Welfer, J. Scharcanski, C. Kitamura, M. D. Pizzol, L. Ludwig, and D. Marinho, “Segmentation of the optic disc in color eye fundus
images using an adaptive morphological approach,” Computers in Biology and Medicine, vol. 40, no. 1, pp. 124–137, 2010.
[17]. Aquino, M. E. Gegundez´-Arias, and D. Mar´ın, “Detecting the optic disc boundary in digital fundus images using morphological,
edge detection, and feature extraction techniques,” Medical Imaging, IEEE Transactions on, vol. 29, no. 11, pp. 1860–1869, 2010.
[18]. Salazar-Gonzalez, Y. Li, and X. Liu, “Optic disc segmentation by incorporating blood vessel compensation,” In Proceedings of
IEEE SSCI, International Workshop on Computational Intelligence In Medical Imaging, pp. 1–8, 2011.
[19]. G. Salazar-Gonzalez, Y. Li, and X. Liu, “Retinal blood vessel segmentation via graph cut,” in Control Automation Robotics &
Vision (ICARCV), 2010 11th International Conference on. IEEE, 2010, 225–230.
[20]. Salazar-Gonzalez, Y. Li, and D. Kaba, “Mrf reconstruction of retinal images for the optic disc segmentation,” in Health Information
Science. Springer, 2012, pp. 88–99.
[21]. G. S. Gonzalez, “Structure analysis and lesion detection from retinal fundus images,” Ph.D. dissertation, Brunel University, 2011.
[22]. Wu, M. Zhang, and J. Liu, “On the adaptive detection of blood vessels in retinal images,” IEEE Transactions on Biomedical
Engineering, vol. 53, no. 2, pp. 341–343, 2006.
[23]. Y. Y. Boykov and M.-P. Jolly, “Interactive graph cuts for optimal boundary & region segmentation of objects in nd images,” in
Com-puter Vision, 2001. ICCV 2001. Proceedings. Eighth IEEE International Conference on, vol. 1. IEEE, 2001, pp. 105–112.
[24]. S. Vicente, V. Kolmogorov, and C. Rother, “Graph cut based image seg-mentation with connectivity priors,” In Proceedings of
IEEE Conference on Computer Vision and Pattern Recognition, CVPR, vol. 1, pp. 1–8, 2008.
[25]. Kolmogorov and Y. Boykov, “What metrics can be approximated by geo-cuts, or global optimization of length/area and flux,” In
Proceedings of Tenth IEEE International Conference on Computer Vision, ICCV., vol. 1, pp. 564–571, 2005.