Generation method and recognition method of section recognition model based on language knowledge guidance

Abstract

The invention discloses a generation method and an recognition method of a section recognition model based on language knowledge guidance. The method comprises the following steps: training a feature extraction model based on a training sample set; determining model parameters of the initial classification model based on the training corpus and the initial graph neural network, and determining a prediction section category and a prediction anatomical structure category based on the feature extraction model and the initial classification model; based on the prediction section category and the prediction anatomical structure category, training the initial image neural network to obtain a classification model; and connecting the feature extraction model and the classification model to obtain a section recognition model. According to the method, the section recognition model is trained through the image features of the training ultrasonic image and the semantic features formed on the basis of the annotation categories carried by the training ultrasonic image, so that the section recognition model can learn the image feature information of the ultrasonic image and the mutual dependency relationship of each section category and each anatomical structure category, and the recognition accuracy based on the section recognition model can be improved.

Description

technical field [0001] The present application relates to the technical field of ultrasonic image processing, and in particular to a generation method and recognition method of a section recognition model guided by language knowledge. Background technique [0002] In routine obstetric screening, ultrasound imaging has been widely used to assess fetal growth and development and congenital malformations. During ultrasound screening, physicians need to select standard slices that include key anatomical structures for measurement of relevant biological parameters, such as figure 1 As shown, the horizontal cross-section of the upper abdomen of the prenatal ultrasound includes anatomical structures such as the spine, gastric vesicles, and umbilical veins (see figure 1 Left panel), abdominal circumference that can be used to assess fetal weight based on the transversal plane of the epigastrium on ultrasound needs to be measured in a standard transversal plane of the upper abdomen....

AI Technical Summary

Problems solved by technology

However, the existing standard sections carrying anatomical structures are generally manually identified by the operator, which is heavily dependent on the operator's experience, so that the identified standard sections vary from operator to operator, thereby affecting the quality of the identified standard sections. accuracy

Images