A rapid screening method, system, device and medium for intracranial aneurysm

By training the semantic segmentation model and combining point cloud data processing, rapid automatic detection and three-dimensional reconstruction of intracranial aneurysms in CTA images are achieved, solving the problems of low diagnostic efficiency and missed detection in the existing technology, and improving the ability to screen asymptomatic populations.

CN119741273APending Publication Date: 2025-04-013201 HOSPITAL
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411810562.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-12-10
Publication Date
2025-04-01

AI Technical Summary

Technical Problem

The prior art is inefficient and difficult to generalize to screen in healthy people when diagnosis of intracranial aneurysms based on CTA images, especially asymptomatic patients with aneurysms are prone to missed examination.

Method used

By obtaining training samples of CTA images of multiple patient heads, the preset initial semantic segmentation model is used for segmentation, and the second segmentation mask is obtained in combination with manual segmentation. The point cloud data is extracted by isosurface extraction method, and the total loss function is calculated to train the target semantic segmentation model. The model is able to automatically detect intracranial aneurysms and reconstruct its three-dimensional surface to measure morphological parameters.

Benefits of technology

Rapid automatic detection and three-dimensional reconstruction of intracranial aneurysms are realized, diagnostic efficiency is improved, and dependence on expert resources is reduced, making it possible to screen aneurysms for asymptomatic high-risk groups.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119741273A_ABST
    Figure CN119741273A_ABST
Patent Text Reader

Abstract

The present invention discloses a method, system, device and medium for rapid screening of intracranial aneurysms, and relates to the technical field of aneurysm screening. The method comprises: calculating a total loss function using a first segmentation mask, a second segmentation mask, first point cloud data and a second point cloud data corresponding to each CTA image, until the total loss function reaches a training end condition, and determining an initial semantic segmentation model that reaches the training end condition as a target semantic segmentation model; acquiring head CTA data of a person to be tested, and inputting the head CTA data into a target semantic segmentation model for processing, obtaining a target segmentation mask corresponding to the head CTA data, and when it is determined that an aneurysm exists in the CTA data, obtaining target morphological parameters according to the target segmentation mask; not only can intracranial aneurysms be automatically detected from CTA images, but also the three-dimensional surface of the aneurysm body can be reconstructed and morphological parameter measurement can be automatically completed, thereby greatly accelerating the clinical workflow.
Need to check novelty before this filing date? Find Prior Art