Surgical navigation system based on visual analysis

By using technical means such as adaptive non-local mean filtering, multi-modal image acquisition, weighted decision tree classification and progressive image registration in the surgical navigation system, the problems of poor image quality, single viewing angle, imbalance in classification and inaccurate registration in the existing system are solved, and the accuracy and safety of surgical navigation are significantly improved.

WO2025059978A1 Publication Date: 2025-03-27CENT SOUTH UNIV
5 Cites 4 Cited by

Patent Information

Application Number
PCT/CN2023/120339
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-09-21
Publication Date
2025-03-27

AI Technical Summary

Technical Problem

The existing surgical navigation system based on visual analysis has problems such as poor image quality, single viewing angle, imbalance in classification and inaccurate registration, resulting in a decrease in the accuracy and safety of surgical navigation.

Method used

Adaptive non-local mean filtering and denoising treatment is used to improve image quality, multi-modal medical images are collected to construct multi-view models, and a weighted decision tree is used for classification, and a medical registration method from coarse to fine combination of progressive images and accelerated robust feature algorithms is used.

Benefits of technology

Provide clear image maps to fully reflect the location of the lesions, improve classification accuracy and efficiency, enhance registration accuracy, thereby improving the overall accuracy of surgical navigation.

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

Abstract

Disclosed is a surgical navigation system based on visual analysis, which comprises an image acquisition module, a data preprocessing module, a data classification module, a registration module, a three-dimensional reconstruction module, and a tracking feedback module. The present invention belongs to the technical field of surgical navigation, relates in particular to a surgical navigation system based on visual analysis, and aims to solve the technical problems of poor quality of an image provided by a three-dimensional model constructed by a medical image in a single mode and a provided angle of view being singular and unable to fully reflect the position of a lesion. According to the present solution, adaptive non-local mean filtering denoising processing is carried out on images, and a multi-modal medical image is acquired for constructing a model having different angles of view and physical information. Meanwhile, according to the present solution, a coarse-to-fine medical registration method which combines a progressive image and a speeded-up robust feature algorithm is used, so that the precision of registration is improved.
Need to check novelty before this filing date? Find Prior Art