Plane wave beam forming method and system based on double-regression convolutional neural network

Abstract

The invention discloses a plane wave beam forming method and system based on a double-regression convolutional neural network, and the method comprises the steps: collecting and preprocessing a multi-angle plane wave echo signal: collecting the multi-angle plane wave echo signal, preprocessing a single-angle plane wave echo signal, and obtaining a radio frequency signal cube; model training: taking a single-angle plane wave radio frequency signal cube as input, taking multi-angle plane wave composite data based on a delay superposition algorithm as a label, and training a pre-constructed double-regression convolutional neural network by using a stochastic gradient descent method; model prediction: taking a single-angle plane wave radio frequency signal cube as input, and predicting data after multi-angle plane wave beam synthesis based on the trained double-regression convolutional neural network; and obtaining a plane wave image through the steps of signal demodulation, logarithm compression and coordinate transformation. The invention also provides a system for implementing the method. According to the invention, the plane wave imaging quality is improved under the condition thatthe frame rate is not reduced.

Description

technical field [0001] The invention belongs to the field of medical ultrasound imaging, and in particular relates to a plane wave beam forming method and system based on a double regression convolutional neural network. Background technique [0002] Ultrafast plane wave imaging is a kind of ultrafast ultrasound imaging. Its essence is that in the process of ultrasonic emission by the transducer, non-focused plane wave emission is used instead of a large number of focused emission to obtain ultrasonic image data. Theoretically, ultrafast plane wave imaging can reach 15,000 frames per second within an imaging depth of 5cm. This imaging method breaks the frame rate limitation of the traditional ultrasound imaging method (100 frames per second), and satisfies many new imaging modes, such as ultrafast blood flow imaging and shear wave elastography. However, due to the use of full-aperture transmission and reception for plane wave imaging, the signal-to-noise ratio of the recei...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a plane wave beamforming method and system based on a double regression convolutional neural network to reduce the possible model function space in view of the problems in the above-mentioned prior art that the possible function mapping space of plane wave imaging is large and the model training is difficult. , effectively enhance the robustness of the model, and improve the plane wave imaging quality as much as possible without reducing the frame rate

Images