Methods and systems for correcting projection data
AssigneeShanghai United Imaging Healthcare Co., Ltd.
InventorYanyan Liu
Life of this patent
Estimating the term…
Abstract
The present disclosure provides a method and system for correction projection data. The method includes: obtaining the entity projection data and the transformed projection data related to the imaging object, the transformed projection data includes the first projection data related to the pixel shading generated by the detector components; inputting the entity projection data, the transformed projection data to the trained correction model and obtaining the correction projection data. The correction model obtained through training realizes the efficient and accurate correction of multi-energy imaging/spectral radiography, so that the obtained correction projection data may be more in line with the complexity of the actual system, and the correction effect may be better.
Claims20
A method, implemented on a computing device having one or more processors and one or more storage devices, comprising:
- obtaining raw projection data by scanning an object using an imaging device including a detector;
- determining, based on the raw projection data, transformed projection data, wherein the transformed projection data includes at least one of:
- first projection data relating to radiation obstructed by the imaging device, or a portion thereof, or
- second projection data relating to a position of the detector, or a portion thereof;
- obtaining corrected projection data by processing the raw projection data and the transformed projection data using a correction model, wherein the correction model includes a trained machine learning model;
- and
- generating a medical image of the object by reconstructing the corrected projection data.
A system for correcting projection data, comprising:
- at least one storage device including a set of instructions or programs;
- and
- at least one processor configured to communicate with the at least one storage device, wherein when executing the set of instructions or programs, the at least one processor is configured to cause the system to perform operations including:
- obtaining raw projection data by scanning an object using an imaging device including a detector;
- determining, based on the raw projection data, transformed projection data, wherein the transformed projection data includes at least one of:
- first projection data relating to radiation obstructed by the imaging device, or a portion thereof,
- second projection data relating to a position of the detector, or a portion thereof, or
- at least one term of an N-th power of a polynomial representation of the raw projection data, wherein N is an integer greater than or equal to 2;
- obtaining corrected projection data by processing the raw projection data and the transformed projection data;
- and
- generating a medical image of the object by reconstructing the corrected projection data.
A non-transitory computer-readable storage medium embodying a computer program product, the computer program product comprising instructions configured to cause a computing device to:
- obtaining raw projection data-by scanning an object using an imaging device including a detector;
- determining, based on the raw projection data, transformed projection data, wherein the transformed projection data includes at least one of:
- first projection data relating to radiation obstructed by the imaging device, or a portion thereof, or
- second projection data relating to a position of the detector, or a portion thereof;
- obtaining corrected projection data by processing the raw projection data and the transformed projection data using a correction model, wherein the correction model includes a trained machine learning model;
- and
- generating a medical image of the object by reconstructing the corrected projection data.
Dependencies are inferred from the claim text.