Abstract
Purpose :
Optical coherence tomography angiography (OCTA) is a promising optical imaging modality that enables label-free 3D blood perfusion imaging in vivo within the tissue bed down to the capillary level. The motion-contrast OCTA is prone to the system noise and environmental fluctuation, which lead to a limited flow contrast. Our study is aimed to develop high flow contrast OCTA.
Methods :
In OCTA, the motion contrast is extracted by statistically classifying dynamic and static scattering signals. The larger sample size, the higher classification reliablity. Here, we propose an approach named ‘maximum available sample size OCTA (MASS-OCTA)’. MASS-OCTA is capable of high sensitivity detection of small flow movement using complex decorrelation analysis by conbining both amplitude and phase information. The sample size is effectively increased using a parallel collection strategy that combines temporal, angular and spectral diversities, and the classification error rate between dynamic and static signals is decreased by integrating multiple independent observations. Furthermore, a hybrid contrast assessment of motion and morphological features is utilized for suppressing false positive flow signals (i.e., static voxels exhibit motion artifacts), which arise from system noise and environmental disturbance in static zones.
Results :
The experimental results have demonstrated that OCTA samples from the temporal, angular or spectral diversity are independent. The classification error rate between dynamic and static signals can be decreased to 1% by integrating a large number of independent samples in MASS-OCTA. Besides, the motion noise in static zones can be further suppressed by using the tubular shape feature of blood flow.
Conclusions :
The proposed MASS-OCTA enables a superior flow contrast for better interpretation and quantification of angiograms.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.