Segmentation method for cerebellar earthworm part in ultrasonic image

Abstract

The invention discloses a segmentation method for a cerebellar earthworm part in an ultrasonic image, and the method comprises the following steps: S1, inputting an ultrasonic image of the cerebellarearthworm part and the size and position information of corpus callosum in the ultrasonic image, and carrying out the preprocessing of the ultrasonic image of the cerebellar earthworm part; S2, observing an ultrasonic image of the cerebellar earthworm part according to medical prior knowledge, and judging direction information of the cerebellar earthworm part in the image; S3, according to the direction information obtained in the step S2 and the morphological characteristics of the cerebellar earthworm part, conducting outer contour graph fitting on the cerebellar earthworm part, and obtaining the initial contour of the cerebellar earthworm part; S4, performing algorithm traumatic injury and smoothing processing on the initial contours of the cerebellar earthworms obtained in the S3 to form precise contours of the cerebellar earthworms; S5, carrying out smoothing processing on the precise contour; and finally, the purpose of completely, accurately and smoothly segmenting the cerebellar earthworm part is achieved.

Description

technical field [0001] The invention relates to the field of medical image processing, in particular to a method for segmenting the cerebellar vermis in an ultrasound image. Background technique [0002] With the widespread application of computer technology in medicine, the earlier manual segmentation of images is gradually replaced by computer automatic segmentation. At present, how to detect whether the fetal cerebellar vermis is developing normally is usually done by manually segmenting the cerebellar vermis region in the ultrasound image, and calculating its contour length, area and other characteristic information to detect whether it is developing normally. How to automatically segment the cerebellar vermis and extract the position and contour information becomes the key issue. At present, the existing cerebellar vermis ultrasound image segmentation methods have problems such as poor accuracy, excessive dependence on ultrasound image quality, poor anti-noise ability,...

AI Technical Summary

Problems solved by technology

At present, the existing ultrasound image segmentation methods of cerebellar vermis have problems such as poor accuracy, over-reliance on ultrasound image quality, poor anti-noise ability, and inaccurate edge segmentation.

Images