Ultrasonic artery segmentation method, system and equipment based on positioning knowledge driving

By employing a priori learning model for localization and a guided adaptive fusion model, the problem of ambiguity in the localization of arterial lumens in ultrasound images was solved, achieving efficient and accurate arterial segmentation and improving the model's stability and data adaptability.

CN121147243AActive Publication Date: 2025-12-16AEROSPACE INFORMATION RES INST CAS
17 Cites 1 Cited by

Patent Information

Application Number
CN202511688069.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-11-18
Publication Date
2025-12-16
Estimated Expiration
2045-11-18

AI Technical Summary

Technical Problem

Existing technologies are unable to effectively solve the problem of ambiguity in the localization of arterial lumens in ultrasound images, resulting in insufficient segmentation accuracy and robustness.

Method used

A localization knowledge-driven approach is adopted, which generates coarse region mask mapping through a localization prior learning model and combines it with a guided adaptive fusion model and a segmentation network to achieve explicit localization guidance and segmentation of ultrasound images.

Benefits of technology

It significantly improves the accuracy and robustness of ultrasound artery segmentation, reduces data annotation costs, enhances model training efficiency and generalization ability, and has stronger interpretability and flexibility.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention provides an ultrasonic artery segmentation method, system and device based on positioning knowledge driving, and relates to the technical field of medical image processing. The method comprises the following steps: acquiring an original ultrasonic image, wherein the original ultrasonic image is an image containing an arterial lumen; on the basis of a positioning prior learning model, rough region mask mapping of the original ultrasonic image is realized, so that a positioning map containing an artery highlight region is obtained; inputting the original ultrasonic image and the positioning map into a fusion model to obtain a fusion feature map; based on a segmentation network model, segmenting the fusion feature map to obtain a target segmentation image; wherein the fusion model is a model for guiding adaptive fusion. According to the method, the problem of fuzzy positioning of the arterial lumen in the ultrasonic image is effectively solved, and correct positioning of the target is ensured, so that the subsequent segmentation is more stable and accurate, and missed segmentation and wrong segmentation are effectively avoided.
Need to check novelty before this filing date? Find Prior Art