An OCT image-based bifurcated blood vessel automatic identification method

Abstract

The invention relates to an OCT image-based bifurcated blood vessel automatic identification method. The method comprises the steps of: firstly acquiring an IVOCT image in a guide wire pullback periodby using an optical coherence tomography system and an angiography system; secondly, determining a blood vessel bifurcation and determining a bifurcation position via a detected blood vessel contour;thirdly, displaying the bifurcated blood vessel. According to the embodiments, bifurcation pairs are acquired through multiple times of screening, and then bifurcation determination is performed andfalse positive is removed through a special method, so that the judgment accuracy is improved, bifurcated blood vessels can be detected and marked rapidly and accurately, and the working efficiency isincreased.

Description

Technical field [0001] The invention belongs to the medical image processing field and the medical detection technical field, and specifically relates to an OCT image-based automatic identification method for bifurcated blood vessels. Background technique [0002] Before stent implantation, accurate measurement of the vascular lumen is very important, which determines the selection of stent size and the optimal position of stent placement. If the size of the stent is not appropriate, it may lead to poor stent adherence. The selection of the optimal position for stent implantation not only needs to consider the smallest lumen area, but also the bifurcated vessels. If a bifurcated vessel is covered by a stent, there is a high risk of pathological changes, such as restenosis and obstruction of blood flow to the bifurcated vessel. Therefore, before stent implantation, it is very important to detect bifurcation vessels and measure main vessels. [0003] Because of its relatively high ...

AI Technical Summary

Problems solved by technology

However, the algorithm relies heavily on the detection of shadows in the binary image, and shadows depend on the threshold setting in the binary image, so the robustness of the algorithm is poor

[0005] In addition, in other existing technologies, the guide wire, catheter, and vessel contour need to be detected before bifurcation detection, and these steps need to use Dijkstra's algorithm separately, which is inefficient; and its judgment on false positives is not comprehensive , the misjudgment rate is high

Images