Label-free vascular network imaging using adaptive Niblack analysis and PIV approach
Kurochkin M.A.1,2, Fedosov I.V.2, Postnov D.E.2, 1skolkovo institute of science and technology, 2Saratov State University.
Abstract
The main functions of the blood microcirculatory system are to provide the traffic of oxygen, nutrients and signal molecules between different parts of the body, as well as the maintenance of water-salt balance. Control of the amount of delivered substances and fluids is carried out at the macro level due to cardiac activity and locally by endothelial cells, which is cover the inner surface of the vessels which can produce the blood viscosity factors as well as vasoconstrictor and vasodilator substances. That is why the structural and functional features of the microcirculation network directly depend on the functional state of the endothelium, which means that it is impossible to adequately assess the state of the blood stream network without its complex analysis of morphological and hemodynamic choracteristics, including the section of capillary network.
Standard methods of vascular visualization by averaging the intensity of a series of frames fail with detection of single erythrocytes running through the capillaries, because by averaging the intensity of the tissue area where only a few frames were present low-intensity object (erythrocyte) leads to an even greater underestimation of its intensity. A significant improvement of the PIV method was carried out in 2019 due to the change from the correlation of zones to the correlation of the corresponding groups of pixels. However, the problem of visualization of the smallest capillaries has not been solved because the proposed filtration methods did not solve the problem of visualization of vessels with a small number of scatterers.
Particle image velocimetry (PIV) method is a cost effective full-field and label-free method for blood cells dynamics analysis based on the calculation of correlation function of the relevant sections of a sequential pair of frames. This makes it possible to accurately analyze the dynamics of micro-flows in the smallest capillaries. To this day, PIV methods for the analysis of blood flows are not complete and require further improvement.
In this work we present the method of adaptation of the PIV analysis method for blood flow visualization and velocity measurement in the capillary network with a low number of red blood cells using the method of frame-by-frame filtration of vascular images via advanced Niblack binarization approach. The proposed method implies the reconstruction of the morphological map of the blood network, including capillaries and the adaptive positioning of the measuring zones of the PIV analysis along this map. The vascular network map was reconstructed by accumulating information about the position of red blood cells contrasting with extended Niblack filtration on each frame of the series.
Speaker
Maxim A. Kurochkin
1Skolkovo institute of science and technology, 2Saratov State University.
Russia
Discussion
Ask question
