Arteriovenous image reconstruction method based on CNN and multi-electrode electromagnetic measurement

Abstract

The invention discloses an arteriovenous image reconstruction method based on CNN and multi-electrode electromagnetic measurement. The arteriovenous image reconstruction method comprises the steps of1, obtaining a data set; 2, dividing a large amount of sample data into a training set and a verification set by adopting ten-fold cross validation; 3, designing a network structure of the CNN; 4, dividing the data into a plurality of batches by adopting a small-batch gradient descent algorithm, and updating parameters in batches; 5, defining a forward propagation process of the CNN; 6, compilinga training module program; 7, compiling a verification module program; 8, carrying out programming through MATLAB software, and calling a meshgrid function and a griddata function; changing respectivecorresponding pixel point values of the original image and the reconstructed image into images; and 9, obtaining distribution of arteries and veins according to an image reconstruction result. According to the invention, based on the convolutional neural network CNN and the multi-electrode electromagnetic measurement theory, the arteriovenous image reconstruction of human limbs is carried out, the regional weight function is determined, and the multi-electrode electromagnetic measurement is better applied to the blood flow velocity measurement of different individuals.

Description

technical field [0001] The invention relates to an arteriovenous image reconstruction method based on CNN and multi-electrode electromagnetic measurement, which belongs to the field of image data processing. Background technique [0002] Blood flow rate is an important indicator for the diagnosis of coronary artery stenosis, coronary heart disease and other cardiovascular diseases. In modern medical methods, the measurement of blood flow velocity can prevent and control in advance the stubborn diseases that have plagued human beings for a long time. Existing studies have shown that based on the electromagnetic flowmeter weight function theory, the multi-electrode electromagnetic measurement method can be applied to the measurement of blood flow velocity in arteries and veins of human limbs. Specific steps are as follows: [0003] 1. Using the finite element analysis method, use the COMSOL Multiphysics simulation software to establish a multi-electrode limb blood measuremen...

AI Technical Summary

Problems solved by technology

However, in actual measurement, the position, size and shape of arteries and veins cannot be determined

Due to differences in human body characteristics, it is even more impossible to determine the regional weight functions under different simulation models for different individuals, which will inevitably affect the measurement accuracy

Images