Real-time image registration and path planning system in bronchoscope surgical navigation

By monitoring and correcting the deformation caused by respiration attraction in real time, the impact assessment, benchmark deformation generation, residual deformation learning and closed-loop correction modules are adopted to solve the navigation error problems caused by respiratory movement during bronchoscopy surgery, achieving higher precision deformation modeling and surgical safety.

CN120477940AActive Publication Date: 2025-08-15THE FIRST AFFILIATED HOSPITAL OF MEDICAL COLLEGE OF XIAN JIAOTONG UNIV
5 Cites 4 Cited by

Patent Information

Application Number
CN202511000695.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-07-21
Publication Date
2025-08-15
Estimated Expiration
2045-07-21

AI Technical Summary

Technical Problem

During bronchoscopic surgery, the patient's respiratory movement and deformation lead to a significant decrease in the registration accuracy of the navigation system, and traditional methods cannot effectively deal with errors caused by respiratory phase changes.

Method used

The impact assessment module, the reference deformation generation module, the residual deformation learning module and the fusion and closed-loop correction module are adopted to monitor and correct the deformation attracted by the call in real time and dynamically adjust the path planning through Pearson correlation coefficient, extended Kalman filtering and multi-layer timing transformation network.

Benefits of technology

It significantly improves the individual accuracy of deformation modeling, can accurately capture local residual deformation and chronic changes, and ensures the accuracy and safety of the surgery.

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

Abstract

The invention discloses a real-time image registration and path planning system in bronchoscope surgical navigation, relates to the technical field of image registration, and is used for solving the problems that in a thoracic and abdominal minimally invasive interventional operation, organ movement and deformation caused by breathing of a patient can cause obvious influence on the registration precision of a navigation system, and the navigation precision is poor. Space registration of preoperative images and an anatomical structure of a patient in an operation often has errors due to mismatching of breathing phase changes; multi-attitude data are fused, the respiration interference degree is comprehensively evaluated through double indexes, the problem that a traditional static deformation field cannot be matched with dynamic respiration is solved through respiratory cycle driving deformation prediction, displacement errors are corrected through tidal volume observation values, and B spline control point weights are reversely optimized through registration residual errors. A deformation field can adapt to breathing amplitude and frequency differences of different patients, online distillation learning is carried out by taking a registration error output by a filter as a teacher signal, and local residual deformation and chronic changes which are difficult to describe by a traditional global model are accurately captured.
Need to check novelty before this filing date? Find Prior Art