As the core organ of cardiovascular system, fetal heart plays a very important role. With the development of pregnancy, the early fetal heart rate tends to mature, and the corresponding cardiac function tends to mature and stable. But the fetal heart is very fragile during pregnancy.
Various diseases during pregnancy directly lead to fetal heart growth restriction, and even lead to fetal heart function damage. Ultrasound image diagnosis is one of the most important diagnostic methods in medical imaging. It is of great significance to detect the early heart structure of
the fetus in gynecology and obstetrics. It can detect the early fetal heart in real time and noninvasively. However, the traditional ultrasonic image detection has many disadvantages in the process of application, such as many noise points, low performance of processing algorithm, which to
some extent affects the detection performance of ultrasonic image detection in the detection of fetal heart structure in gynecology and obstetrics. Based on the above problems, this paper proposes an adaptive detection algorithm of superimposed moving image based on ultrasonic image detection,
which can accurately extract and analyze fetal heart region when the signal-to-noise ratio of ultrasonic image sequence is low. The average anisotropy algorithm is also proposed innovatively in this paper. In order to predict the structure of fetal heart more accurately, the active heart model
combining fetal heart structure and motion information is considered in the actual analysis process. Experiments show that the accuracy error of the algorithm is less than 11 pixels.