Patent · US Active

Method for optimizing ultrasonic imaging system parameter based on deep learning

US11564661B2 · kind B2 · utility

0Cited by
0References
7Claims
0Family size

Assignee

Inventors

Key dates

Filing dateJun 29, 2018
Grant dateJan 31, 2023
Priority date
Expiry dateMay 2, 2039

Classification

  • Technology area (CPC G)Physics
  • CPC primaryG06T2207/20084
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

A method for optimizing an ultrasonic imaging system parameter based on deep learning, comprising the following steps: step 1: collecting samples for training neural networks, the samples comprising ultrasound image samples i, and a corresponding ultrasonic imaging system parameter vector sample p used by an ultrasonic imaging system when the ultrasonic image samples are collected; step 2: establishing a neural network model and training the neural networks to convergence by using the samples collected=in step 1, so as to obtain a trained neural network system onn; and step 3: taking the original ultrasonic imaging system parameter vector p or the original ultrasonic image as an input to be input into the neural network system onn trained in step 2, at this moment, a parameter obtained from an output end of onn being an optimized ultrasonic imaging system parameter vector ep=onn(p). By means of the method, the purpose of improving the ultrasonic image quality is realized by optimizing the ultrasonic imaging system parameter.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.