Method of calibrating x-ray projection geometry in x-ray cone beam computed tomography
A method of x-ray projection geometry calibration in x-ray cone beam computed tomography, including: at least one step (S 1 ) of obtaining two-dimensional x-ray images or a sinogram of at least a part of an object, generated through relatively rotating around the object a detector and an x-ray source projecting x-rays towards the detector; further including: at least one step (S 4 ) of detecting in the two dimensional x-ray images or the sinogram at least one feature of the object by using a trained artificial intelligence algorithm; and at least one step of creating, based on the detection, calibration information which defines the geometry of the x-ray projection.
1. A method of x-ray projection geometry calibration in x-ray cone beam computed tomography, the method comprising:
obtaining a sinogram of at least part of an object, wherein the sinogram is generated through relatively rotating around the object, a detector and an x-ray source projecting x-rays towards the detector;
detecting in the sinogram, at least one feature of the object by using a trained artificial intelligence algorithm, wherein
the object is a patient, and
the at least one feature of the object includes an anatomical part or a prosthetic of a body or a head of the patient;
creating, based on the detected sinogram, calibration information which defines a geometry of a x-ray projection, wherein the created calibration information corresponds to a patient-specific calibration information; and
reconstructing, based on the sinogram of the at least part of the body of the patient and the corresponding patient-specific calibration information, a three-dimensional tomographic image.
2. The method of claim 1 , wherein the trained artificial intelligence algorithm generates a 3D mask which represents a 3D location and a 3D shape of the feature of the object in the sinogram.
3. The method of claim 1 , wherein the anatomical part comprises at least part of a tooth or a bone, and the prosthetic part comprises at least part of an implant.
4. The method of claim 1 , further comprises:
storing a recently created patient-specific calibration information, or
updating a previously stored patient-specific calibration information with the recently created patient-specific calibration information respectively.
5. The method of claim 1 , further comprises:
pre-processing the sinogram before detecting the at least one feature of the object in the sinogram.
6. The method of claim 5 , wherein the pre-processing the sinogram comprises at least one of a process of filtering, a contrast enhancement, an edge enhancement, and a noise suppression.
7. The method of claim 1 , further comprises:
selecting one or more a features to be detected in the sinogram.
8. The method of claim 1 , further comprises:
training the artificial intelligence algorithm by using a plurality of data pairs, wherein each data pair of the plurality of data pairs includes a sinogram and an associated 3D mask which represents a 3D location and a 3D shape of a feature in the sinogram.
9. A system for x-ray cone beam computed tomography, comprising:
an x-ray unit comprising an acquisition means adapted to acquire a sinogram of at least part of an object through relatively rotating, around the object, an x-ray source and a detector;
a tomographic reconstruction unit comprising an image processing means adapted to detect in the sinogram, at least one feature of the object by using a trained artificial intelligence algorithm, wherein
the object is a patient, and
the at least one feature of the object includes an anatomical part or a prosthetic of a body or a head of the patient;
create, based on the detected sinogram, calibration information which defines a geometry of a x-ray projection, wherein the created calibration information corresponds to a patient-specific calibration information; and
reconstruct, based on the sinogram of the at least part of the body of the patient and the corresponding patient-specific calibration information, a three-dimensional tomographic image.
10. A non-transitory computer-readable medium storing a program in the medium comprising codes for causing a computer-based x-ray volume tomography system to perform a method of:
obtaining a sinogram of at least part of an object, wherein the sinogram is generated through relatively rotating around the object, a detector and an x-ray source projecting x-rays towards the detector;
detecting in the sinogram, at least one feature of the object by using a trained artificial intelligence algorithm, wherein
the object is a patient, and
the at least one feature of the object includes an anatomical part or a prosthetic of a body or a head of the patient;
creating, based on the detected sinogram, calibration information which defines a geometry of a x-ray projection, wherein the created calibration information corresponds to a patient-specific calibration information; and
reconstructing, based on the sinogram of the at least part of the body of the patient and the corresponding patient-specific calibration information, a three-dimensional tomographic image.