Methods and systems for improving 3D volume reconstruction from 2D X-ray images
Inventors
HOERNDLER KLAUS, null • Hörndler, Klaus • KRAENZEL WOLFGANG, null • Kränzel, Wolfgang • Fleischmann, Christof
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-7712961-B2
Publication Date
2010-05-11
Expiration Date
Abstract
A method of improving the volume reconstruction in navigationally guided operations is provided. In certain implementations, a phantom which contains X-ray positive marks and marks detectable by a position detection system in a fixed spatial relation to each other is positioned in the region of the volume being investigated on the subject. For the volume reconstruction, a series of 2D X-ray projection images can be created, each of them containing, alongside the structures of the subject being examined, the images of the X-ray positive marks of the phantom as image information. Each 2D data set is transformed by a familiar method, in which, for example, the mechanical deviations of a real-world C-arm type X-ray diagnostic machine are identified by a kinematic model and are corrected. The transformed 2D data can be used for the volume reconstruction, and the volume model with the position information of the X-ray positive marks is relayed to a navigation system.
Core Innovation
The invention relates to a method for improving volume reconstruction in navigationally guided operations on a subject of investigation using an X-ray diagnostic machine. A phantom containing X-ray positive marks is arranged on the subject so that the phantom remains rigidly connected to the subject during the entire operation and remains fixed in space during each scan. A series of 2D projection images is acquired with different projection geometries, where each projection contains at least some of the X-ray positive marks.
From the 2D projection images obtained from each scan, the positions of the X-ray positive marks on the phantom are determined in the image processing computer. Those determined positions are compared with calculated nominal positions of the X-ray positive marks for each scan as determined by the position detection system. This comparison is used to calibrate and transform each 2D projection image into a transformed 2D data set by minimizing a sum of deviations between the phantom mark positions determined from the projection and the phantom mark positions in the transformed 2D data set.
The transformation additionally divides any deviation of the 2D projection of the phantom from the value calculated from kinematics equally between the X-ray receiver and the X-ray source. A volume is then reconstructed with the transformed 2D data sets. The approach relays the phantom-referenced reconstructed X-ray volume to a navigation system, referenced to a coordinate system connected to the phantom’s X-ray positive marks.
Claims Coverage
The document provides one independent claim. It defines rigid phantom fixing, acquisition of 2D projections at different projection geometries, marker position determination, calibration and transformation by comparison to nominal positions from a position detection system, a kinematics-based deviation allocation, and volume reconstruction from the transformed 2D data sets.
Rigidly connected X-ray phantom during operation
Arranging the phantom on the subject in a manner such that the phantom remains rigidly connected to said subject during the entire operation and remains fixed in space during each scan.
Acquiring 2D projection images with varying projection geometries
Using the X-ray diagnostic machine to create a series of 2D projection images of the region being explored with different projection geometries, each projection containing at least some of the X-ray positive marks of the phantom.
Determining phantom marker positions from 2D projections
Determining the positions of the X-ray positive marks on the phantom from the 2D projection images obtained from each scan in the image processing computer.
Calibrating and transforming projections via marker deviation minimization and kinematics-based allocation
Comparing the positions of the X-ray positive marks on the phantom with the calculated nominal positions of the X-ray positive marks for each scan determined by the position detection system, calibrating and transforming each 2D projection image to a transformed 2D data set by minimizing the sum of deviations, and dividing any deviation of the 2D projection of the phantom from the value calculated from kinematics equally between the X-ray receiver and the X-ray source.
Reconstructing volume from transformed 2D data sets
Reconstructing a volume with the transformed 2D data sets.
Relaying phantom-referenced volume to navigation system
Relaying the reconstructed X-ray volume to a navigation system referenced to a coordinate system connected to the phantom’s X-ray positive marks.
Overall, the independent claim covers improved 3D volume reconstruction by rigidly fixing a phantom with X-ray positive marks to the subject, determining marker positions from multiple 2D projections, and using a position detection system to compute calibrated transformations before reconstructing the volume and relaying it to a navigation system.
Stated Advantages
Improving volume reconstruction in navigationally guided operations.
Documented Applications
Navigationally guided operations on a subject of investigation using an X-ray diagnostic machine and a navigation system.
Interested in licensing this patent?