Ultrasonic image recognition method, terminal equipment and storage medium

Abstract

The invention discloses an ultrasonic image recognition method, terminal equipment and a storage medium. The method comprises the steps of obtaining a probability mapping feature map corresponding toa to-be-recognized ultrasonic image; determining and obtaining a plurality of candidate feature points corresponding to the ultrasonic image based on the probability mapping feature map; and updatingthe probability mapping feature map based on the plurality of candidate feature points, and determining a target area corresponding to the ultrasonic image based on the updated probability mapping feature map. According to the method, the probability mapping feature map of the ultrasonic image is acquired, a plurality of candidate feature points are corrected based on the probability mapping feature map to update the probability mapping feature map, and finally ovarian follicles in the ultrasonic image are segmented. In this way, the accuracy of the target area determined based on the probability mapping feature map can be improved, and therefore doctors can better observe and analyze the development situation of ovarian follicles.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to an ultrasonic image recognition method, a terminal device and a storage medium. Background technique [0002] At present, clinically, doctors often use ultrasound to observe the size and volume of women's ovaries during the ovulation cycle, as well as the development of follicles, to predict the formation period of mature oocytes, thereby improving the success rate of reproduction. However, in the process of statistical analysis, doctors need to perform a large number of frequent manual operations to calculate the number of follicles, measure the maximum meridian and vertical bisector of follicles, etc., especially in patients who have been treated with drugs, there are generally multiple follicles to be tested in the body, manual measurement more cumbersome. Moreover, follicle monitoring is real-time and continuous, and human errors introduced due to different ...

AI Technical Summary

Problems solved by technology

However, in the process of statistical analysis, doctors need to perform a large number of frequent manual operations to calculate the number of follicles, measure the maximum meridian and vertical bisector of follicles, etc., especially in patients who have been treated with drugs, there are generally multiple follicles to be tested in the body, manual measurement more complicated

Moreover, follicle monitoring is real-time and continuous, and human error introduced due to different habits of manual marking and measurement by different doctors will affect the final monitoring effect

Images