A three-dimensional blood flow imaging method and system based on feature space

Abstract

The invention discloses a three-dimensional blood flow imaging method and system based on characteristic space. OCT scattered signals of scattered signal samples inside the three-dimensional space arecollected through a collector; a classifier is combined with the OCT scattered signals and the decorrelation for constructing a two-dimensional characteristic space, and classification of dynamic blood flow signals and static tissue is achieved. The method comprises the specific steps that time decorrelation is adopted for calculating the OCT scattered signals to analyze relative movement of blood flow and peripheral tissue, and OCTA blood flow information of the OCT scattered signals is obtained; intensity information and OCTA blood flow information of the OCT scattered signals are combined,the intensity-decorrelation coefficient two-dimensional characteristic space is constructed, and the OCT scattered signals are projected into the space; the OCT scattered signals are classified in the constructed classifier space; according to the classification result of the classifier, a blood flow radiographic map is generated. The problem that an existing single-threshold intensity film has the unideal inhibitory effect for artifact caused by random noise can be solved, and the blood flow image quality is improved.

Description

technical field [0001] The present invention relates generally to the field of biomedical imaging, and more particularly to blood flow imaging and machine learning methods associated with Optical Coherence Tomography (OCT) and blood flow imaging (OCT Angiography, OCT-A) . Background technique [0002] Blood perfusion is an important index to measure physiological function and pathological state. Currently, the commonly used angiography technique in clinical practice requires intravenous injection of exogenous markers, which may cause side effects that make it unsuitable for long-term and frequent blood flow monitoring. Track detection. In recent years, OCTA, an angiographic technique developed on the basis of optical coherence tomography, replaces traditional exogenous fluorescent markers with endogenous blood flow movement. The ability of clear and reliable three-dimensional imaging of the microvascular network of the microvascular network has made this technology develop...

AI Technical Summary

Problems solved by technology

However, simple intensity masks lead to high classification error rates and low motion contrast due to complex dependencies between decorrelation coefficients and signal intensity

[0005] Some existing OCTA classification methods will be interfered by the difference between the phantom model and the actual biological tissue, and cannot clearly distinguish the accurate classification of dynamic and static signals

Images