|
|
Bimonthly Since 1986 |
ISSN 1004-9037
|
|
|
|
|
Publication Details |
Edited by: Editorial Board of Journal of Data Acquisition and Processing
P.O. Box 2704, Beijing 100190, P.R. China
Sponsored by: Institute of Computing Technology, CAS & China Computer Federation
Undertaken by: Institute of Computing Technology, CAS
Published by: SCIENCE PRESS, BEIJING, CHINA
Distributed by:
China: All Local Post Offices
|
|
|
|
|
|
|
|
|
|
Abstract
Medical image processing is a tool and technique for creating a visual image of inside of the body. The rapid advancement of digital imaging and computer vision has broadened the potential for the use of imaging technology in medicine. Image processing is especially useful in diagnostic medical systems. Reliable glaucoma detection in digital fundus images remains an open problem in biomedical image processing. Detection of glaucoma in the retinal fundus image is necessary to avoid loss of vision. The early detection of glaucoma and elevated inter ocular pressure is critically significant to arrest the progression of the irreversible disease. With the ocular degeneration of the retina and optic nerve, and other ailments such as high blood sugar levels, chronic blood pressure, thyroid imbalance and cataract, manual detection of glaucoma is often challenging leading to significant vision loss at the time of detection. Moreover, at low income group countries, medical costs also act as a deterrent for the early detection of the disease causing permanent vision loss. Hence, off late, research on the development of automated tools has been of paramount importance which can detect glaucoma with low false negative and false positive rates. In this paper, an image enhancement and feature extraction technique is developed which is subsequently used to train a deep neural network. Gray Level Co-occurrence matrix based features along with the cup to disc ratio have been used to train a Deep Bayes Net with regularization for automated detection of glaucoma. The performance of the proposed system has been evaluated in terms of the classification accuracy of the system.
Keyword
Fundus Image, Optic Disc, Biomedical image processing, Optic Disc detection, Bayesian Regularization, Cup to Disc Ratio (CDR).
PDF Download (click here)
|
|
|
|
|