US-12620093-B2 - Method and apparatus for allocating image processing

US12620093B2US 12620093 B2US12620093 B2US 12620093B2US-12620093-B2

Abstract

The present application provides a method, a non-transitory computer-readable storage medium, and apparatus for allocating image processing. The method for allocating image processing can include, on the basis of relevant information of a medical image, selecting, according to an allocation list, a learning network model corresponding to the information from a plurality of learning network models. The method can also include performing image processing on the medical image on the basis of the selected learning network model, the allocation list including a list of correspondences between relevant information of medical images and the plurality of learning network models.

Inventors

Huayi Wang
Haibo Mei
Jingliang Tian
Tao Wang

Assignees

GE Precision Healthcare LLC

Dates

Publication Date: 20260505
Application Date: 20230922
Priority Date: 20220922

Claims (18)

1 . A method for allocating image processing, comprising: on the basis of relevant information of a current medical image, selecting, according to an allocation list, a learning network model corresponding to the relevant information from a plurality of learning network models, the allocation list comprising a list of correspondences between relevant information of medical images and the plurality of learning network models, wherein, in response to detecting that a piece of relevant information has no corresponding learning network model in the allocation list, creating a correspondence between the relevant information and a learning network model in the allocation list on the basis of a preset rule by: providing the medical image as input into each learning network model for image processing; for each processed image generated as output by a learning network model that is evaluated as successful, creating a temporary correspondence between the relevant information and the learning network model, and incrementing a count; and in response to the count of the temporary correspondence reaching a preset value, creating a new correspondence between the relevant information and the learning network model in the allocation list; and performing image processing on the current medical image on the basis of the selected learning network model.
2 . The method for allocating image processing according to claim 1 , wherein the relevant information of the medical image comprises at least one among an imaging device type, a scan site and a possible disease type.
3 . The method for allocating image processing according to claim 2 , wherein acquisition of the scan site comprises image recognition of the medical image or extraction of header file information of the medical image.
4 . The method for allocating image processing according to claim 2 , wherein the allocation list comprises correspondences between imaging device types and scan sites, and the plurality of learning network models, and selecting the learning network model comprises selecting a corresponding learning network model on the basis of the imaging device type and the scan site.
5 . The method for allocating image processing according to claim 1 , wherein the plurality of learning network models comprises at least two among a number n of CT models, a number m of MR models, and a number t of X-ray models, the n CT models being configured to perform image processing on n scan sites in a CT image, respectively, the m MR models being configured to perform image processing on m scan sites in an MR image, respectively, and the t X-ray models being configured to perform image processing on t scan sites in an X-ray image, respectively, n, m and t being integers.
6 . The method for allocating image processing according to claim 1 , wherein the plurality of learning network models are used to perform at least one among image segmentation, image optimization and image rendering on an image.
7 . The method for allocating image processing according to claim 1 , further comprising: determining whether the relevant information of the medical image has a corresponding learning network model in the allocation list.
8 . The method for allocating image processing according to claim 1 , wherein the preset rule comprises that a processed image outputted by any learning network model is evaluated as successful more than a preset number of times.
9 . The method for allocating image processing according to claim 8 , wherein whether the processed image is successful is evaluated on the basis of at least one among image quality evaluation, site matching degree and sequence integrity evaluation.
10 . A non-transitory computer-readable storage medium, which is used to store a computer program that, when executed by a computer, causes the computer to: on the basis of relevant information of a current medical image, select, according to an allocation list, a learning network model corresponding to the relevant information from a plurality of learning network models, the allocation list comprising a list of correspondences between relevant information of medical images and the plurality of learning network models, wherein, in response to detecting that a piece of relevant information has no corresponding learning network model in the allocation list, creating a correspondence between the relevant information and a learning network model in the allocation list on the basis of a preset rule by: providing the medical image as input into each learning network model for image processing; for each processed image generated as output by a learning network model that is evaluated as successful, creating a temporary correspondence between the relevant information and the learning network model, and incrementing a count; and in response to the count of the temporary correspondence reaching a preset value, creating a new correspondence between the relevant information and the learning network model in the allocation list; and perform image processing on the current medical image on the basis of the selected learning network model.
11 . The non-transitory computer-readable storage medium for allocating image processing according to claim 10 , wherein the relevant information of the medical image comprises at least one among an imaging device type, a scan site and a possible disease type.
12 . The non-transitory computer-readable storage medium for allocating image processing according to claim 11 , wherein acquisition of the scan site comprises image recognition of the medical image or extraction of header file information of the medical image.
13 . The non-transitory computer-readable storage medium for allocating image processing according to claim 11 , wherein the allocation list comprises correspondences between imaging device types and scan sites, and the plurality of learning network models, and selecting the learning network model comprises selecting a corresponding learning network model on the basis of the imaging device type and the scan site.
14 . The non-transitory computer-readable storage medium for allocating image processing according to claim 10 , wherein the plurality of learning network models comprises at least two among a number n of CT models, a number m of MR models, and a number t of X-ray models, the n CT models being configured to perform image processing on n scan sites in a CT image, respectively, the m MR models being configured to perform image processing on m scan sites in an MR image, respectively, and the t X-ray models being configured to perform image processing on t scan sites in an X-ray image, respectively, n, m and t being integers.
15 . The non-transitory computer-readable storage medium for allocating image processing according to claim 10 , wherein the plurality of learning network models are used to perform at least one among image segmentation, image optimization and image rendering on an image.
16 . An image processing apparatus, comprising: a plurality of learning network models used to perform image processing on a medical image; a memory used to store an allocation list comprising a list of correspondences between relevant information of medical images and the plurality of learning network models; and a processor configured to: on the basis of relevant information of a medical image, select, according to the allocation list, a learning network model corresponding to the relevant information from the plurality of learning network models, wherein, in response to detecting that a piece of relevant information has no corresponding learning network model in the allocation list, creating a correspondence between the relevant information and a learning network model in the allocation list on the basis of a preset rule by: providing the medical image as input into each learning network model for image processing; for each processed image generated as output by a learning network model that is evaluated as successful, creating a temporary correspondence between the relevant information and the learning network model, and incrementing a count; and in response to the count of the temporary correspondence reaching a preset value, creating a new correspondence between the relevant information and the learning network model in the allocation list; and perform image processing on the medical image on the basis of the selected learning network model.
17 . The image processing apparatus for allocating image processing according to claim 16 , wherein the relevant information of the medical image comprises at least one among an imaging device type, a scan site and a possible disease type.
18 . The image processing apparatus for allocating image processing according to claim 17 , wherein acquisition of the scan site comprises image recognition of the medical image or extraction of header file information of the medical image.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS The present matter claims priority to Chinese Patent Application 202211155704.7, filed Sep. 22, 2022, the contents of which are incorporated by reference herein. TECHNICAL FIELD The present invention relates to medical image processing, and in particular, to a method, a non-transitory computer-readable storage medium, and apparatus for allocating image processing. BACKGROUND In general, medical imaging devices comprise computed tomography (CT) systems, magnetic resonance imaging (MRI) systems, X-ray imaging systems and the like, all of which are capable of acquiring medical images. However, in order to assist diagnosis, image processing, e.g., image segmentation, image post-processing and the like, of the images is usually required. In general, after a medical image is acquired, image processing of the medical image is performed directly in the medical imaging device. However, with increasing demand for assisted diagnosis, the amount of computing and the time required for image processing by utilizing medical imaging devices are both gradually increasing, such that the usage rate of the imaging devices has been reduced, and the number of patients scanned per day has also decreased. SUMMARY The present invention provides a method, a non-transitory computer-readable storage medium, and apparatus for allocating image processing. An exemplary embodiment of the present invention provides a method for allocating image processing, comprising: on the basis of relevant information of a medical image, selecting, according to an allocation list, a learning network model corresponding to the information from a plurality of learning network models, the allocation list comprising a list of correspondences between relevant information of medical images and the plurality of learning network models; and performing image processing on the medical image on the basis of the selected learning network model. An exemplary embodiment of the present invention further provides a non-transitory computer-readable storage medium, which is used to store a computer program that, when executed by a computer, causes the computer to execute instructions for the method for allocating image processing described above. An exemplary embodiment of the present invention further provides an image processing apparatus, the image processing apparatus comprising: a plurality of learning network models, a memory, and a processor, the plurality of learning network models being used to perform image processing on a medical image, the memory being used to store an allocation list, the allocation list comprising a list of correspondences between relevant information of medical images and the plurality of learning network models, the processor being used to, on the basis of the relevant information of the medical image, select, according to the allocation list, a learning network model corresponding to the information from the plurality of learning network models, and perform image processing on the medical image on the basis of the selected learning network model. Other features and aspects will become apparent from the following detailed description, drawings, and claims. BRIEF DESCRIPTION OF THE DRAWINGS The present invention can be better understood by means of the description of the exemplary embodiments of the present invention in conjunction with the drawings, in which: FIG. 1 is a schematic diagram of an operating environment according to some embodiments of the present invention; FIG. 2 is a schematic diagram of an image processing apparatus according to some embodiments of the present invention; FIG. 3 is a schematic diagram of an operation procedure for image processing according to some embodiments of the present invention; FIG. 4 is a schematic diagram of header file information of a medical image according to some embodiments of the present invention; FIG. 5 is a schematic diagram of an allocation list according to some embodiments of the present invention; FIG. 6 is a flowchart of a method for allocating image processing according to some embodiments of the present invention; and FIG. 7 is a flowchart of a method for allocating image processing according to some other embodiments of the present invention. DETAILED DESCRIPTION Specific embodiments of the present invention will be described below. It should be noted that in the specific description of said embodiments, for the sake of brevity and conciseness, the present description cannot describe all of the features of the actual embodiments in detail. It should be understood that in the actual implementation process of any embodiment, just as in the process of any engineering project or design project, a variety of specific decisions are often made to achieve specific goals of the developer and to meet system-related or business-related constraints, which may also vary from one embodiment to another. Furthermore, it should also be understood that alth