Paper Info
Title
Blind Source Separation Of Different Retinal Pulsatile Patterns From Simultaneous Long-Term Binocular Ophthalmoscopic Video-Records
Abstract
Optical imaging of retinal hemodynamic function is an important part of ophthalmologic research. Development and inventing of imaging devices and data analysis methods are both just in progress. The current study innovatively implements two blind source separation (BSS) techniques (i.e. spatial Principal Component Analysis - sPCA; and spatial Independent Component Analysis - sICA) in application of an automatic detection and segmentation of a distinct Optic Disc (OD) areas with different hemodynamic properties from a simultaneous binocular video-ophthalmoscopic records. Both methods detected 3 different spatial patterns mostly symmetric over both eyes stable and reproducible over investigated participants, i.e. central Spontaneous Vessel Pulsations (SVPs), inner OD intensity pulsations and other OD pulsations. Dynamics of all mentioned patterns has a periodic character with similar main frequency (possibly corresponding to subject-specific heart rate) but shifted phase decreasing patterns' mutual high cross correlations. The sICA estimates a higher rate of phase shifts than sPCA.
Year
DOI
Venue
2019
10.1109/EMBC.2019.8857560
2019 41ST ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY (EMBC)
Field
DocType
Volume
Computer vision,Pulsatile flow,Segmentation,Computer science,Retina,Optic disc,Artificial intelligence,Retinal,Optical imaging,Blind signal separation,Principal component analysis
Conference
2019
ISSN
Citations 
PageRank 
1557-170X
0
0.34
References 
Authors
0
7
Name
Order
Citations
PageRank
Ivana Labounková101.01
René Labounek200.34
Jan Odstrcilík331.12
Michal Hracho400.68
Igor Nestrasil500.34
Ralf-Peter Tornow600.68
Radim Kolár701.35