Medical ultrasonic image segmentation method and device

Abstract

The invention provides a medical ultrasonic image segmentation method and device, and the method comprises the steps: obtaining a human body ultrasonic image through ultrasonic scanning equipment; training the semantic segmentation model to obtain a semantic segmentation prediction model; segmenting the to-be-detected image according to the semantic segmentation prediction model; and performing optimization processing on the segmented to-be-detected image to obtain a final target area. The medical ultrasonic image segmentation method and device have the advantages that a deep learning method and a traditional image processing method are combined, the problem that segmentation is inaccurate when the signal-to-noise ratio of the ultrasonic image is low can be effectively solved, the segmentation precision of the bladder ultrasonic image can be improved, and the anti-interference performance of the algorithm is improved.

Description

technical field [0001] The invention relates to the field of image segmentation, in particular to a medical ultrasound image segmentation method and device. Background technique [0002] Image segmentation plays an important role in imaging diagnosis. Automatic segmentation can help doctors quickly determine the size and location of organs or lesions, and can reduce the heavy workload caused by manual labeling by doctors. [0003] The bladder is a muscular cystic organ that stores urine in the human body. The volume of the bladder reflects the amount of urine that the human body can store, and is also an important parameter for clinical application in urology. In the diagnosis and treatment of the urinary system, the measurement results of the residual urine volume of the human bladder and the filled urine volume of the bladder are important references for doctors to diagnose cases. Non-invasive measurement of bladder volume using ultrasound equipment has been widely used c...

AI Technical Summary

Problems solved by technology

[0004] In the prior art, there have been many bladder image segmentation methods, but the existing bladder ultrasound image segmentation, because there are many artifacts in the bladder ultrasound image, the anterior wall of the bladder often has brighter multiple reflection signals, resulting in a simple The threshold-based segmentation method is difficult to completely segment the bladder area

Region-based segmentation methods are also affected by this artifact and lead to inaccurate segmentation

In addition, in bladder ultrasound images, on the bladder side wall, due to the incident angle and attenuation of the ultrasound signal, the image signal-to-noise ratio is very low, and the algorithm based on edge detection is not ideal for segmentation in such a low signal-to-noise ratio image

Based on the method of deep learning, the accuracy of segmentation is relatively high, but the segmentation accuracy is not high enough, especially the segmented bladder boundary, which often cannot completely fit the edge of the bladder wall

Images