Automated classification of glaucoma stages using higher order cumulant features

Noronha, Kevin; Acharya, U. Rajendra; Nayak, K. Prabhakar; Martis, Roshan Joy; Bhandary, Sulatha V.

doi:10.1016/j.bspc.2013.11.006

Cited by 145 publications

(55 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In case of normal eye CDR ranges from 0.2 to 0.5 [12] but in case of the glaucoma, It may reaches up to 1 [13]. Since vertically oval OD and horizontally oval cup, the CDR in normal eye is considerably larger horizontally than vertically so the quotient of the H-V CDR is usually higher than 1.0 [12].…”

Section: Real Time Classificationmentioning

confidence: 93%

Detection of optic disc and cup from color retinal images for automated diagnosis of glaucoma

Lotankar¹,

Noronha²,

Koti³

2015

2015 IEEE UP Section Conference on Electrical Computer and Electronics (UPCON)

View full text Add to dashboard Cite

Glaucoma is the major cause of ocular damage and vision loss in which increased Intraocular Pressure (IOP) of the eye progressively damages the optic nerve. In this proposed study, an automatic system is developed for glaucoma detection by extracting various features like vertical Cup to Disc Ratio (CDR), Horizontal to Vertical CDR (H-V CDR), Cup to Disc Area Ratio(CDAR), and Rim to Disc Area Ratio (RDAR) from digital fundus images through segmentation of Optic Disc (OD), cup and neuroretinal rim. OD is segmented using Geodesic active contour model and cup is detected using color information of the pallor region in M channel of CMY color space. The performance evaluation of the proposed technique has been carried out on 150 images comprising 75 normal and 75 glaucoma images using a set of supervised classifiers namely Naïve Bayes(NB), Support Vector Machine (SVM), and kNearest Neighbor (k-NN). On the private database, the proposed system yields the highest accuracy, Positive Predictive Value (PPV), Negative Predictive Value (NPV), specificity and sensitivity of 99.22%, 84.41%, 86.30%, 84% and 86.66% respectively using k-NN classifier. The results obtained by proposed technique indicate that this glaucoma detection system is beneficial for the clinicians in glaucoma screening programs.

show abstract

Section: Real Time Classificationmentioning

confidence: 93%

Detection of optic disc and cup from color retinal images for automated diagnosis of glaucoma

Lotankar¹,

Noronha²,

Koti³

2015

2015 IEEE UP Section Conference on Electrical Computer and Electronics (UPCON)

View full text Add to dashboard Cite

show abstract

“…We compare the performance of our proposed method based on the accuracy achievement. The accuracy achieved by [21] is 87.5%, [8] is 71.2%, [17] is 88%, [12] is 90% and [10] is 92.65%.…”

mentioning

confidence: 99%

“…Subsequently, the feature extraction of the ONH based on shape proposed by using moment method in the image of Confocal Scanning Laser Tomography (CSLT) [7], Self-Organizing Map (SOM) [8] and Variational Expectation Maximization (VEM) on the image of Heidelberg Retina Tomograph (HRT) [9]. Several feature extraction methods of the texture approached by using Higher Order Spectra (HOS) method [10], the combination of HOS and Discrete Wavelet Transform (DWT) [11][12], wavelet sub-bands (wavelet subbands Daubechies (Db4), Symlets (sym4) and filter Biorthogonal (bi03.7, bi04.2 & bi04.7)) [13]. Fractal Dimension [14], Gray Level Co-occurrence Matrix (GLCM) [15].…”

Section: Introductionmentioning

confidence: 99%

“…Furthermore, there are several methods used to the classification phase such as: Support Vector Machines (SVM) [7], [11], [16][17], Naïve Bayesian [10], statistical analysis [14], Multilayer Perceptron (MLP) [7], [21], Probabilistic Neural Network (PNN) [12], Learning Vector Quantization (LVQ) [8] and the combination of MLP with back propagation [13].…”

Section: Introductionmentioning

confidence: 99%

“…It is based on the comparison of the results with several previous reaserces which use different feature extraction method as shown in Table 2. Based on table 2, glaucoma detection produced by several approaches, such as Blood vessel extraction using SOMs [21],Vector features extraction using SOM [8], Raw intensities, fourier analysis and B-spline interpolation [17], DWT method [12] and HOS cumulants [10]. We compare the performance of our proposed method based on the accuracy achievement.…”

mentioning

confidence: 99%

See 2 more Smart Citations

The Contour Extraction of Cup in Fundus Images for Glaucoma Detection

Septiarini¹,

Hamdani²,

Khairina³

2016

IJECE

View full text Add to dashboard Cite

Glaucoma is the second leading cause of blindness in the world; therefore the detection of glaucoma is required. The detection of glaucoma is used to distinguish whether a patient's eye is normal or glaucoma. An expert observed the structure of the retina using fundus image to detect glaucoma. In this research, we propose feature extraction method based on cup area contour using fundus images to detect glaucoma. Our proposed method has been evaluated on 44 fundus images consisting of 23 normal and 21 glaucoma. The data is divided into two parts: firstly, used to the learning phase and secondly, used to the testing phase. In order to identify the fundus images including the class of normal or glaucoma, we applied Support Vector Machines (SVM) method. The performance of our method achieves the accuracy of 94.44%. Keyword: Contour features Cup Fundus image Glaucoma Morphology Copyright © 2016 Institute of Advanced Engineering and Science.All rights reserved. Corresponding Author:Anindita Septiarini, Faculty of Computer Science and Information Technology, Universitas Mulawarman, Samarinda-75123, Indonesia. Email: anindita@unmul.ac.id INTRODUCTIONGlaucoma is an eye disease that can cause vision problems. This disease is the second cause of blindness in the world [1], therefore the detection of glaucoma is required. The detection of glaucoma by an expert carried out by observing the structure of the retina using a fundus image, to determine the eyes of the patient are normal and glaucoma. Currently, research on automatic glaucoma detection techniques have been developed. Most of the research using fundus images as input data and several feature extraction methods have been applied.Several previous researches related to automatic glaucoma detection generated by applying nonmorphological feature extraction method (image-based features). The feature referred to color, shape or texture on the optic nerve head (ONH), retinal nerve fiber layer (RNFL) or blood vessel as part of the structure of the retina [2]. These parts undergo changes in patients with glaucoma [3]. ONH (also called disc) is the part area with round shape. Inside of the disc, there is a smaller area with round shape called a cup and a neuroretinal rim which is an area that lies between the disc and the cup. RNFL is the area outside of the disk where the structure of thickness of the RNFL is the one which distinguishes whether a patient's eye is normal or glaucoma. The illustration of ONH and RNFL structure are shown in Figure 1.In the features extraction of the ONH applied the process of blood vessel inpainting as the previous phase in order to remove or blurring the blood vessel that presences as noise. This process is required the detection to determine the presence of the blood vessel. Several methods have been proposed for the detection of the blood vessel. Thresholding with the low pixel values in the image followed by median filtering of the size 41x41 [4], using threshold probing [5] and normalized mathematical morphology (used to enhance the vessel...

show abstract

Deep learning in medical imaging: A brief review

Serte¹,

Serener²,

Al‐Turjman³

2020

Trans Emerging Tel Tech

View full text Add to dashboard Cite

Researchers have used deep learning methods for a human level or better disease identification and detection. This paper reports, in brief, the recent work in deep learning identification of diseases occurring at three unique parts of the human body: the skin, the thorax, and the eye. While earlier reviews reported on the theory, applications, and challenges of such research, what distinguishes this work from the others is the reporting and comprehensive analysis of the key results. In doing so, the paper not only summarizes the major conclusions drawn from them but also emphasizes their weaknesses. The hope is to help the researchers see the big picture in deep learning classification of the diseases of the skin, the thorax and the eye, and guide them to find the right future research direction. INTRODUCTIONPhysicians use different imaging modality techniques, such as MRI, CT, ultrasound, and X-ray to help them detect, diagnose, and cure serious diseases. With help from radiologists, they can interpret a vast number of diseases occurring at distinct parts of a human body, ranging from brain to lungs to skin. Still, the tremendous diversity in disease pathology can make human errors possible.In search of computer methods to aid in disease detection and recognition, researchers have experimented on convolutional neural networks (CNN) since the mid-90s. 1 Used for hand-written digit recognition at first, 2 CNN's popularity increased with the discovery of deep CNNs in 2012. [3][4][5][6][7][8] Since then, a growing number of researchers are using CNNs for medical image analysis (such as References 9-17). Some even show that CNNs can give comparable performance to medical doctors. 18 We may use the Internet of Things (IoT) to connect medical devices in hospitals for the collection of big data. Each hospital can join a central unit this way for the automated detection of diseases on images using CNN. IoT is a crucial tool for reaching data in an instant, and artificial intelligence is essential for modeling for disease detection on images. Together they may form what we call The Internet of Medical Things (IoMT). 19 CNN's distinct architectures and vast application areas (such as this and References 20-22) have likewise led the research groups to publish a few review papers on using deep CNNs in medical imaging (eg,. While these papers focus on the theory, applications, and challenges of deep CNNs, they do not detail the results discovered by the researchers.This survey paper, yet, aims to fill that gap by reporting the recent advances of deep learning use in medical imaging and analyzing the key results. In doing that, it hopes to offer the reader an opportunity to see the big picture, including any strengths and weaknesses that such research might have. The intention is not to cover the entire application areas in medical imaging but focus on only a few parts of the human body and be thorough. Hence, this survey paper includes

show abstract

Automated classification of glaucoma stages using higher order cumulant features

Cited by 145 publications

References 41 publications

Detection of optic disc and cup from color retinal images for automated diagnosis of glaucoma

Detection of optic disc and cup from color retinal images for automated diagnosis of glaucoma

The Contour Extraction of Cup in Fundus Images for Glaucoma Detection

Deep learning in medical imaging: A brief review

Contact Info

Product

Resources

About