Ultrasonic imaging method, device and ultrasonic equipment thereof
An ultrasonic imaging method and image technology, applied in the fields of ultrasonic/sonic/infrasonic Permian technology, ultrasonic/sonic/infrasonic image/data processing, organ movement/change detection, etc., can solve cumbersome and time-consuming problems, and achieve The effect of increasing accuracy and reducing operational complexity
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0031] like figure 1 As shown, the present invention provides an ultrasonic imaging method for automatically counting and marking the vertebrae in the image, the method comprising:
[0032] S101. Acquire a complete fetal spine image.
[0033] Collect a complete volume of fetal spine data through the probe to obtain a clear image of the fetal spine. The image can be in 2D, 3D mode.
[0034] 2D: The probe emits ultrasonic signals to the tissue, the echoes such as reflection and scattering produced by organs and tissues with different impedances and different attenuation characteristics are received by the probe, and electrical signals are generated after the ultrasonic piezoelectric effect. After signal processing, the processed The digital scan conversion and image post-processing are performed on the data, and the ultrasonic two-dimensional image of the tissue section is displayed.
[0035] 3D: Use a mechanical motor or manually drive the probe to move, acquire a series of ...
Embodiment 2
[0044] In some embodiments, the step S102 extracting each vertebra image from the fetal spine image may specifically include:
[0045] S1021. Extract an image of a fetal spine region from the fetal spine image.
[0046] On the image of the fetal spine, the doctor marks the range of the fetal spine that needs to be counted by a rectangular frame or a free trace.
[0047] S1022. Perform binarization processing on the extracted fetal spine region image.
[0048] Before the binarization process, the threshold needs to be determined. There are many methods for determining the threshold, such as p-quantile method, histogram concave analysis method, maximum inter-class variance method, entropy method, minimum error threshold, etc. A suitable threshold can be found using . Take the widely used method of maximum between-class variance as an example.
[0049] For an image I(x, y) of size M×N, set the segmentation threshold of foreground and background as T. Under the condition that ...
Embodiment 3
[0086] In some embodiments, the step S103 counting and marking each vertebra image includes:
[0087] According to the connected domain results obtained, the total number of connected domains is the number of recognized vertebrae, and each connected domain is labeled for the doctor's reference to obtain the final result.
[0088] Further, the step S103 counts and marks each of the spine images, and after the above-mentioned connected domain label, it also includes: establishing a recognition template according to the prior knowledge of the spine structure, and correcting the final vertebrae according to the template. Count results.
[0089] like Image 6 As shown, since each vertebra contains two or three connected regions in the 3D image on some slices, in the calculated connected region, it is possible that the three parts of the same vertebra will have different connectivity Therefore, in order to increase the accuracy of the calculation, we can also establish a recogniti...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


