Method and device for determining registration reliability of intravascular ultrasound image

By calculating the confidence metric of intravascular ultrasound images to assess the reliability of image registration, the problem of amplified mismatch errors in IVUS images is solved, thereby improving the robustness of image processing and diagnostic accuracy.

CN122453884APending Publication Date: 2026-07-24SONOSEMI MEDICAL CO LTD
0 Cites 0 Cited by

Patent Information

Application Number
CN202610678870.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-05-18
Publication Date
2026-07-24

AI Technical Summary

Technical Problem

In the existing IVUS image processing workflow, the accuracy of image registration is insufficient, which leads to the amplification of mismatch errors during spatial composite process, affecting the accuracy of clinical diagnosis.

Method used

By acquiring reference and target images from intravascular ultrasound, various confidence metrics (such as global image alignment error, uniqueness of suboptimal registration results, and local sharpness) are calculated, and a comprehensive confidence score and mismatch discrimination label are determined based on these metrics to evaluate the reliability of image registration.

Benefits of technology

It improves the robustness of IVUS multi-frame image processing and the reliability of the output composite image, reduces blurring, ghosting and tissue distortion, and improves the accuracy of clinical diagnosis.

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

Abstract

The invention provides a method and a device for determining registration reliability of an intravascular ultrasound image. The method comprises the following steps: acquiring an intravascular ultrasound reference image and a target image after image registration; determining a first confidence measure evidence, a second confidence measure evidence, and a third confidence measure evidence based on the reference image and the target image; the first confidence degree measurement evidence is used for representing alignment error minimization based on a global image, the second confidence degree measurement evidence is used for representing global uniqueness based on a suboptimal registration result, and the third confidence degree measurement evidence is used for representing local sharpness based on discrete curvature; a comprehensive confidence score and a mismatch discrimination tag are determined based on the first confidence metric evidence, the second confidence metric evidence, and the third confidence metric evidence. Whether the image registration result of the reference image and the target image is reliable or not is judged through various confidence degree measurement evidences, and the robustness of IVUS multi-frame image processing and the reliability of the output composite image are improved.
Need to check novelty before this filing date? Find Prior Art