Method of radiography of an organ of a patient

Inventors

Desaute, Pascal • METZ, Camille

Assignees

EOS Imaging SA

Interested in licensing this patent?

MTEC can help explore whether this patent might be available for licensing for your application.

Publication Number

US-11864940-B2

Patent

Publication Date

2024-01-09

Expiration Date


Abstract

Disclosed is a method of radiography of an organ of a patient, including: first and second vertical scanning being performed synchronously, wherein a computed correction is processed on both first and second raw images, on at least part of patient scanned height, for at least overweight or obese patients, so as to reduce, between first and second corrected images, cross-scattering existing between the first and second raw images, and wherein the computed correction processing on both the first and second raw images includes: a step of making a patient specific modeling, using as patient specific data therefore at least both first and second raw images, a step of determining a patient specific representation of radiation scattering by the patient specific modeling, a step of processing the patient specific radiation scattering representation on both the first and second raw images so as to get the first and second corrected images.

Core Innovation

The invention relates to radiography of an organ of a patient using first vertical scanning and second vertical scanning performed synchronously by a first radiation source and first detector, and by a second radiation source and second detector. The synchronized scans produce a first two-dimensional raw image and a second two-dimensional raw image viewing the organ according to different angles of incidence. The problem addressed is that, for at least overweight or obese patients, radiation scattering degrades image quality by introducing cross-scattering between the first and second raw images.

To solve this, the invention performs computed correction processing on both the first and second raw images on at least part of data associated with scanned patient height for at least overweight or obese patients to reduce cross-scattering between the first and second corrected images. The computed correction processing includes making a patient-specific model by using at least both of the first and second raw images as patient-specific data, where the model is a three-dimensional (3D) voxelized reconstruction obtained from deformation of a generic model to patient-specific radiographic projections corresponding to the first and second raw images.

The computed correction further includes determining a patient-specific representation of radiation scattering by making the patient-specific model by directly simulating scatter signals scattered by the 3D voxelized reconstructions. Finally, processing the determined patient-specific radiation scattering representation on both the first and second 2D raw images obtains the first and second corrected images. The document also describes scatter modeling using a patient-specific scatter matrix representing radiation scattering obtained by interpolation between a pre-computed base of matrices corresponding to simple geometries, followed by simulation of radiation scattering based on that scatter matrix and processing to obtain corrected images.

Claims Coverage

The document contains three independent claims: clm-00001 (method), clm-00028 (apparatus), and clm-00029 (method with scatter-matrix interpolation). Across these, the inventive subject matter centers on synchronized stereo-vertical scanning from different incidence angles and computed correction to reduce cross-scattering for at least overweight or obese patients using patient-specific scatter modeling derived from both raw images (either via direct scatter simulation on a 3D voxelized reconstruction, or via a patient-specific scatter matrix interpolated from a pre-computed simulated base).

Synchronized vertical scanning for different incidence angles

The method includes first vertical scanning of the organ by a first radiation source and a first detector to make a first two-dimensional raw image, and second vertical scanning of the organ by a second radiation source and a second detector to make a second two-dimensional raw image, the first and second vertical scanning being performed synchronously, the first and second raw images viewing the organ according to different angles of incidence.

Cross-scattering reduction computed correction on scanned patient height for overweight or obese patients

The method performs computed correction processing on both the first and second raw images, on at least part of data associated with scanned patient height for at least overweight or obese patients, to reduce, between the first and second corrected images, cross-scattering existing between the first and second raw images.

Patient-specific 3D voxelized reconstruction and direct scatter simulation

The computed correction processing comprises making a patient-specific model, using at least both of the first and second raw images as patient-specific data, where the patient-specific model is a three-dimensional (3D) voxelized reconstruction obtained from deformation of a generic model to patient-specific radiographic projections corresponding to the first and second raw images; determining a patient-specific representation of radiation scattering by making the patient specific model by directly simulating scatter signals scattered by the 3D voxelized reconstructions; and processing the determined patient-specific radiation scattering representation on both the first and second 2D raw images to obtain the first and second corrected images.

Patient-specific model from voxelized geometry and scatter matrix interpolation

The computed correction processing comprises making a patient-specific model, using at least both of the first and second raw images, where the patient-specific model is a voxelized geometry corresponding to a combination of simple geometries; obtaining a patient-specific scatter matrix representing radiation scattering of the patient-specific model from a pre-computed base of matrices corresponding to the simple geometries by interpolation between matrices among the pre-computed base of matrices, the pre-computed base of matrices being obtained with simulations; determining a patient-specific representation of radiation scattering by simulating the radiation scattering of the voxelized geometry based on the patient-specific scatter matrix; and processing the patient-specific radiation scattering representation on both the first and second raw images to obtain the first and second corrected images.

Overall, the claims cover synchronized acquisition of two 2D raw images of an organ from different incidence angles, followed by computed correction that reduces cross-scattering between views for at least overweight or obese patients. The correction is grounded in patient-specific radiation scattering modeling derived from raw images, either using a deformed generic model to form a 3D voxelized reconstruction with direct scatter signal simulation, or using a patient-specific scatter matrix interpolated from a pre-computed simulated base for scatter simulation and subsequent correction image formation.

Stated Advantages

Reduce cross-scattering between first and second corrected images.

Applicable to at least overweight or obese patients by performing correction on at least part of data associated with scanned patient height.

Documented Applications

Radiography of an organ of a patient, including at least overweight or obese patients, where computed correction processing is used to reduce cross-scattering between views.

JOIN OUR MAILING LIST

Stay Connected with MTEC

Keep up with active and upcoming solicitations, MTEC news and other valuable information.