Handheld three-dimensional scanner-based spatial registration method for surgical navigation

Abstract

The invention belongs to the field of medical equipment, and particularly relates to a handheld three-dimensional scanner-based spatial registration method for surgical navigation. The method adopts facial recognition feature point matching registration to perform preliminary registration on space point clouds of a to-be-tested person obtained by the handheld three-dimensional scanner and image space point clouds obtained by a medical image; a preliminary registration result uses a point cloud iterative registration method to complete the registration of the space of the to-be-tested person ofsurgical navigation and the image space; implementation results show that the method has fast registration speed and stable registration precision, and can meet the requirements of a surgical navigation system and improve the registration accuracy of the surgical navigation system.

Description

technical field [0001] The invention belongs to the field of medical image processing and medical equipment, and relates to a spatial registration method for surgical navigation based on a hand-held three-dimensional scanner. Background technique [0002] The prior art discloses that the surgical navigation system can provide doctors with lesion location and boundary information in clinical practice, which is conducive to guiding and improving the quality of surgery, and has gradually become an indispensable auxiliary device in neurosurgery. The industry believes that the core function of surgical navigation is to register the patient space where the surgical area is located with the image space where the 3D model is located, and then use the registration results to guide the surgical instruments. It can be seen that the performance of the surgical navigation system is largely affected by the registration. impact on accuracy. [0003] At present, the widely used registratio...

AI Technical Summary

Problems solved by technology

[0003] At present, the widely used registration method in surgical navigation is point matching. Its essence is to determine the matching of the entire space through the matching of limited points. Practice shows that there are still some defects in the registration method, for example, the registration accuracy of point matching It cannot represent the overall matching accuracy of the space; the feature points for point matching need to be manually selected, which is prone to errors; an additional image scan is required for navigation; it takes a long time to select feature points, etc.

[0004] The prior art also discloses a registration method related to surface matching. The registration method based on surface matching does not require manual setting of feature points, nor additional image scanning, and can avoid the above-mentioned defects of point matching; moreover, surface matching More feature information is used, which can improve the accuracy of registration. However, the main reason why face matching is still not widely used in clinical practice is that the traditional point cloud acquisition method is difficult to obtain complete facial point cloud data. , and with the increase of facial feature information, the time-consuming growth of registration and other defects

[0005] In order to determine the correlation between two sets of image space point clouds, some studies use the method of finding the correlation between single points, and then apply the obtained correlation to the whole point cloud to obtain an updated point cloud, and then update the point cloud Cloud searches for the correlation of single points, so iteratively, and finally obtains the point cloud transformation matrix that is close to the optimal solution; the method is intuitive and simple, and has high registration accuracy, but it is still sensitive to the initial position, and it is easy to fall into a local optimum. Excellent but unable to reach the global optimal defect

Images