Automatic identification method and system for fetal standard section in ultrasonic image

Abstract

The invention belongs to the technical field of prenatal ultrasonic examination. The invention further discloses an automatic identification method of the fetal standard section in the ultrasonic image. The method comprises the following steps: obtaining a data set, performing preprocessing operation on the obtained data set to obtain a preprocessed data set, and inputting the obtained preprocessed data set into a trained deep convolutional neural network to obtain the category and position of a target contained in each fetal ultrasound section image. The deep convolutional neural network comprises a backbone network ResNet-50, a feature pyramid network, a classification sub-network and a positioning sub-network which are connected in sequence. According to the invention, the technical problems of quite high technical requirements on doctors, relatively low accuracy and consistency caused by differences of examination results of different doctors and overlong examination time in an existing standard section identification method can be solved.

Description

technical field [0001] The invention belongs to the technical field of prenatal ultrasonic examination, and more particularly relates to an automatic identification method and system for a fetal standard slice in an ultrasonic image. Background technique [0002] Ultrasound imaging has been widely used in pregnancy diagnosis due to its advantages of low consumption, real-time imaging and no radiation. In the process of ultrasonic diagnosis of fetal diseases using ultrasound imaging, the sonographer first needs to obtain each standard section of the fetus, and then identify the standard section, check the main anatomical structure through the standard section, and then perform further diagnosis and examination, according to fetal growth The situation and whether the fetal anatomy is abnormal are analyzed and diagnosed. [0003] In the current ultrasonic diagnosis process, the identification of standard slices is mainly obtained by manual marking of points and lines based on ...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the above defects or improvement needs of the prior art, the present invention provides a method and system for automatic identification of fetal standard slices in ultrasound images. High requirements, low accuracy and consistency caused by differences in the examination results of different sonographers, long examination time, frequent operations and labor-intensive technical problems during the examination by sonographers

Images