Target tracking method and device suitable for ultrasound contrast dual-mode dynamic image, medium and product

By employing a dual-path Kalman filter and a novel covariance joint regression method in dual-mode dynamic ultrasound contrast imaging, the problem of the lack of cross-modal information interaction and collaboration mechanism was solved, achieving robustness and accuracy of target tracking in complex clinical scenarios, and ensuring smooth and stable output of target trajectory and automation level.

CN122289319AActive Publication Date: 2026-06-26SHENZHEN UNIV
5 Cites 0 Cited by

Patent Information

Application Number
CN202610627938.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-05-09
Publication Date
2026-06-26
Estimated Expiration
2046-05-09

AI Technical Summary

Technical Problem

Existing ultrasound contrast imaging dual-mode dynamic image target tracking suffers from a lack of cross-modal information interaction and collaboration mechanisms, and insufficient estimation accuracy of the covariance matrix (Q, R) of observation noise and process noise, resulting in poor tracking robustness and difficulty in adapting to complex clinical scenarios such as respiratory motion, probe micro-movement, and lesion deformation.

Method used

A dual-path Kalman filter is used, with each Kalman filter configured independently. Information is coordinated by exchanging the observation matrix, observation vector, and observation noise. The covariance matrix of process noise and observation noise is solved simultaneously by using joint regression of information covariance and recursive least squares method of matrix forgetting factor. Combined with cross-modal information interaction and closed-loop feedback mechanism, deep coordination of dual-mode information is achieved.

Benefits of technology

It improves the robustness and accuracy of target tracking, can adapt to complex clinical scenarios, ensures smooth and stable output of target trajectory, reduces operational complexity, and improves the level of automation.

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

Abstract

The invention discloses a target tracking method and device suitable for an ultrasound contrast dual-mode dynamic image, a medium and a product, and relates to the technical field of target tracking, and the method comprises the steps: respectively configuring a Kalman filter for a dissection mode and a contrast mode; executing state prediction based on the fusion result of the previous frame; extracting a dual-mode observation vector and calculating information and a residual error; synchronously solving and correcting a noise covariance matrix by adopting innovation covariance joint regression and matrix forgetting factor recursive least square; executing double measurement updating and cross-modal information interaction; and carrying out weighted fusion on the posterior state and carrying out closed-loop feedback to the next frame. According to the method, dual-mode deep collaboration, noise covariance adaptive estimation and closed-loop correction can be realized, and the robustness and accuracy of target tracking in a complex clinical scene are remarkably improved.
Need to check novelty before this filing date? Find Prior Art