Abstract Details

Title Automated Segmentation and Multifractal Analysis of High-Resolution Non-Invasive Capillary Perfusion Maps (nCPMs) of the Human Retina

Topic Neuro-ophthalmology/Neuro-otology

Presentation(s) IN4 - (-)

Poster/Presentation
Number 002

Objective

Background The retina provides a window to study the etiology and pathogenesis of cerebrovascular diseases. Pathologic retinal microvascular changes may reflect micro-angiopathic processes in the brain. Previous fractal analysis of the retinal vasculature is limited by its low resolution of the fundus photo for extracting vessel information. Recent advances in optical imaging enable visualization of the retinal microvasculature at the capillary level and construction of high resolution non-invasive capillary perfusion maps (nCPMs).

Design/Methods nCPMs were obtained using the Retinal Function Imager (RFI, Optical Imaging Ltd, Rehovot, Israel). Image segmentation was custom developed. After correcting non-uniform illumination of the image, morphological filters were used for extracting linear shapes which were mostly vessels and non-vessel structures. The grey-scale images were processed using thresholding in order to create the dense network in the nCPMs. Matlab's bwmorph algorithm was used to skeletonize the segmented vascular network. Fractal analysis toolbox from Benoit (TruSoft Benoit Pro 2.0, TruSoft International Inc) was used to analyze the fractal dimension of the nCPMs. Multiple nCPMs of 2 healthy subjects were obtained and analyzed.

Results The nCPM showed much more details of the microvascular network, which appeared to be very dense and balanced branching of the capillary network around the avascular zone (the fovea). The network was well preserved in the segmented image. Multifractal analysis (D0) of the nCPM was performed and the result was 1.92 ± 0.03.

Conclusions This study has demonstrated for the first time that automated segmentation and multifractal analysis of nCPMs are feasible. This approach may open a new era for studying retinal microvasculature and brain small vessel disease.

Authors/Disclosures
Hong Jiang, MD, PhD (University of Miami) PRESENTER	The institution of Dr. Jiang has received research support from Florida Health. The institution of Dr. Jiang has received research support from NINDS RO1.
Robert L. Knobler, MD, PhD (Knobler Institute of Neurologic Disease, PC)	No disclosure on file
	No disclosure on file
Tatjana Rundek, MD, PhD	The institution of Dr. Rundek has received research support from NIH.
	No disclosure on file
	No disclosure on file
	No disclosure on file
	No disclosure on file

��ɫ��

��ɫ��