International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 10 Issue: 04 | Apr 2023
p-ISSN: 2395-0072
www.irjet.net
RETINA DISEASE IDENTIFICATION USING IMAGE PROCESSING Madhavan P1, Istamsetty Siva Krishna2, Shaik Parvez3, Paidi Sathish4 1Assistant Professor, Dept. of ECE, Muthayammal Engineering College, Rasipuram. 2UG Scholar, Dept. of ECE, Muthayammal Engineering College, Rasipuram.
3UG Scholar, Dept. of ECE, Muthayammal Engineering College, Rasipuram.
4UG Scholar, Dept. of ECE, Muthayammal Engineering College, Rasipuram.
--------------------------------------------------------------***-------------------------------------------------------------
Abstract - The use of imaging and computer vision
imaging, on the other hand, includes digital video that can be viewed without special equipment. Diagnostic radiography specifically refers to the technical aspects of medical imaging and the acquisition of medical images. Radiologists are usually responsible for obtaining high-quality medical images for diagnosis, but some radiation treatments can be performed by radiologists.
systems allows for a quantitative study of human physiology. A recent study has developed an algorithm that combines image processing and machine learning techniques to analyze retinal images and aid in the early detection and diagnosis of retinal diseases. The main aim is to apply these techniques to digital fundus images of the eye to accurately separate diseased eyes from normal ones and improve the speed and accessibility of retinal disease diagnosis and treatment. Automated analysis of retinal images is crucial in diagnostic procedures, and the approach presented in this study utilizes datasets of retinal images to classify over 180 fundus images with lesions and non-lesions, achieving an accuracy of 94.4%, a precision of 94%, a recall and f1-score of 94%, and an AUC of 95%. The proposed approach employs image processing and the Support Vector Machine (SVM) classification method to distinguish diseased eyes from normal eyes using fundus images, thus paving the way for precise and automated classification and diagnosis of retinal diseases.
Retinal image processing plays a crucial role in the diagnosis and treatment of various diseases that affect the retina and the choroid. One such disease is diabetic retinopathy, which is a complication of diabetes mellitus that affects the retina and the choroid. The advent of retinal imaging technology has enabled optometrists to capture digital images of the retina, blood vessels, and optic nerve located at the back of the eyes. This has greatly aided in the early detection and management of diseases that can affect both eyes and overall health, such as glaucoma, macular degeneration, diabetes, and hypertension. With retinal imaging technology, even the slightest changes to the structures at the back of the eyes can be detected. For instance, in choroidal neovascularization (CNV), a network of small blood vessels arises in the choroid and takes away a portion of the blood supplying the retina. As a result, the sight may be degraded and, in severe cases, vision loss may occur. Adaptive Optics (AO) has the potential to facilitate early detection of retinal pathologies. Many researchers have been working on retinal images to perform various image processing tasks for the benefit of the health sector. However, the accuracy of the image analysis depends on the quality of the images, which must have high contrast photoreceptors and vasculature, as well as accurate registration.
Key Words: Retinal image processing, arteries, veins, segmentation, classification, identification.
1. INTRODUCTION Medical imaging refers to the techniques and processes of visually representing the internal structures of the body for the purpose of analyzing and intervening in health conditions. We aim to clarify hidden internal structures covered by skin and bones, and to diagnose and treat diseases. By creating a database of normal anatomy and physiology, medical imaging can identify abnormalities in various organs of the body. Imaging of excised organs and tissues can be done for medical reasons, but is generally considered part of pathology rather than medical imaging.
Currently, many researchers have developed methods for automatically assessing the quality of retinal images taken by a fundal camera, using a reference image. Recently, AO has been combined with scanning laser ophthalmoscope and optical coherence tomography (OCT) to obtain images
In a clinical context, medical imaging using "invisible light" is usually associated with radiology or medical imaging, and radiologists are responsible for understanding and sometimes capturing these images. “Visible light” medical
© 2023, IRJET
|
Impact Factor value: 8.226
|
ISO 9001:2008 Certified Journal
|
Page 329