PointNet-based neural navigation fast surface registration method and PointNet-based neural navigation fast surface registration system

Abstract

The invention belongs to the technical field of medical image processing, and particularly relates to a Point Net-based neural navigation fast surface registration method and system. The method comprises the following steps: performing head segmentation on a preoperative image by using a threshold segmentation algorithm, morphological operation and the like, and reconstructing to obtain a three-dimensional surface point set; adopting GO! SCAN three-dimensional handheld scanner to obtain a head surface point set in a real physical space; registering the anatomical mark points of the patient byusing a singular value decomposition method to obtain coarse registration transformation; and obtaining final fine registration transformation of the two sets of point sets through a Point Net-based surface point set rapid registration method. The system comprises four corresponding modules: a head surface reconstruction module, a handheld three-dimensional laser scanner module, a coarse registration module and a fine registration module. The method is high in operation efficiency and reliable in registration precision, can be integrated in an existing neural navigation system, achieves the quick registration of an image space and a real physical space, and improves the positioning precision of surgical navigation.

Description

technical field [0001] The invention belongs to the technical field of medical image processing, and in particular relates to a PointNet-based neuronavigation fast plane registration method and system. Background technique [0002] After the neurosurgery navigation establishes the corresponding relationship between the image space and the physical space where the patient is located, the relative positional relationship between the intraoperative surgical instrument and its surrounding anatomical structures can be displayed on the preoperative image in real time to assist doctors in performing surgery. The most critical technology is to establish the corresponding relationship between the two spaces, that is, neuronavigation space registration technology. The higher the accuracy of the registration, the more accurate the information provided to the physician by the resulting navigation system. In addition, a fast registration process is the basis for a smooth operation. The...

AI Technical Summary

Problems solved by technology

Its disadvantages are: (1) It is necessary to scan an additional image with markers before the operation, which increases the complexity of the operation and the patient's expenditure.

(2) The pasted marker points are prone to move and introduce errors, resulting in a decrease in navigation accuracy

However, due to the large number of point sets participating in the registration, the fine registration algorithms currently used, such as the Iterative Closest Point (ICP) algorithm, the Coherent Point Drift (CPD), and the Gaussian mixture model-based (Gaussian Mixture Model, GMM) registration algorithm, the iterative optimization process of these algorithms is time-consuming, the calculation efficiency is low, and it is difficult to achieve real-time registration of neuronavigation

Images