A New System for Measuring the Auto-Fluo Changes in Age-Related Macula Degeneration after Intravenous Injection of Bavecizumab Medicine
Mohammad Faraji1, Mohammad Norouzi Fard2, Saeed Mirghasemi3

1Mohammad Faraji, Department of Computer & IT Engineering, Islamic Azad University, Parand Branch, Tehran, Iran.
2Mohammad Norouzi Fard, Department of Computer Engineering, Islamic Azad University, Science & Research Branch, Tehran, Iran.
3Saeed Mirghasemi (Corresponding Author), Department of Computer & IT Engineering, Islamic Azad University, Parand Branch, Tehran, Ira.

Manuscript received on 12 December 2012 | Revised Manuscript received on 21 December 2012 | Manuscript Published on 30 December 2012 | PP: 1-6 | Volume-2 Issue-1, December 2012 | Retrieval Number: A0350112112 /2012©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: In aged people, age-related macula degeneration is the second prevalent disease after diabetes which causes blindness. The only cure for age-related macula degeneration is the Bavecizumab intravenous medicine injection. To prove this treatment, the number of dead cells in macula area should be considered. In this paper, to obtain the number of dead cells, a novel system has been presented for measuring the existing auto fluorescence in macula area of retinal images. This combinational system is composed from three parts; pre-processing of retinal, processing the images, and understanding the images. The pre-processing level, includes eliminating margins, and reversing retina image. In processing level, the image is segmented, and features are extracted, where the segmentation has been done using techniques like morphology, dynamic thresholding and connected components. The specifications of target areas are the Euclidian distance to the center of the image, and density. In the understanding level of image, collecting the specifications of each class, macula area and the measurable parameter for evaluating the amount of auto fluorescence is obtained which is useful for determining the number of dead cells in macula area. The results are concluded using probabilistic analysis including linear regression and correlation between data. The method is tested on a database composed of 34 retina images belonging to patients of age-related macula degeneration.
Keywords: Age-related macula degeneration, Connected components, Morphology, Macula, Retina image.

Scope of the Article: Healthcare Informatics