WO2023102749A1

WO2023102749A1 - Image processing method and system

Info

Publication number: WO2023102749A1
Application number: PCT/CN2021/136183
Authority: WO
Inventors: 倪成; 陈伟梁; 李博; 傅费超; 汪鹏; 李娅宁
Original assignee: 上海联影医疗科技股份有限公司
Priority date: 2021-12-07
Filing date: 2021-12-07
Publication date: 2023-06-15
Also published as: CN117940958A

Abstract

Embodiments of the present description provide an image processing method and system. The method comprises: obtaining a target image of an object; obtaining at least one reference image of the object, the imaging device corresponding to the at least one reference image being different from the imaging device corresponding to the target image; and correcting the target image on the basis of the at least one reference image.

Description

A kind of image processing method and system

technical field

This description relates to the field of medical technology, in particular to an image processing method and system.

Background technique

In recent years, medical imaging has been widely used in the diagnosis and treatment of various medical conditions. Different imaging methods have different advantages and disadvantages. For example, images acquired by computed tomography (CT) have high spatial position accuracy, but poor soft tissue resolution; magnetic resonance (MR) images have low spatial position accuracy, but have good soft tissue resolution. Therefore, it is necessary to provide an image processing method and system that can combine the advantages of different imaging methods.

Contents of the invention

One of the embodiments of this specification provides an image processing system. The image processing system includes a storage device and a processor, the storage device is used to store computer instructions; the processor is used to connect with the storage device. When executing the computer instructions, the processor causes the system to perform the following operations: acquire a target image of the object; acquire at least one reference image of the object, the at least one reference image corresponds to a different imaging device an imaging device corresponding to the target image; and correcting the target image based on the at least one reference image.

One of the embodiments of this specification provides an image processing method. The method includes: acquiring a target image of the object; acquiring at least one reference image of the object, the imaging device corresponding to the at least one reference image is different from the imaging device corresponding to the target image; and based on the at least one reference image A reference image is used to correct the target image.

One of the embodiments of the present specification provides a computer-readable storage medium, the storage medium stores computer instructions, and when a computer reads the computer instructions, the computer executes an image processing method. The method includes: acquiring a target image of the object; acquiring at least one reference image of the object, the imaging device corresponding to the at least one reference image is different from the imaging device corresponding to the target image; and based on the at least one reference image A reference image is used to correct the target image.

One of the embodiments of this specification provides an image processing system. The system includes: a target image acquisition module, used to acquire a target image of the object; a reference image acquisition module, used to acquire at least one reference image of the object, and the imaging device corresponding to the at least one reference image is different from the an imaging device corresponding to the target image; and a correction module, configured to correct the target image based on the at least one reference image.

Description of drawings

This specification will be further illustrated by way of exemplary embodiments, which will be described in detail with the accompanying drawings. These examples are non-limiting, and in these examples, the same number indicates the same structure, wherein:

Fig. 1 is a schematic diagram of an application scenario of an exemplary image processing system according to some embodiments of this specification;

Fig. 2 is a block diagram of an exemplary image processing system according to some embodiments of the present specification;

Fig. 3 is a flow chart of an exemplary image processing method according to some embodiments of this specification;

Fig. 4A is a schematic diagram of an exemplary reference image according to some embodiments of the present specification;

Fig. 4B is a schematic diagram of an exemplary target image according to some embodiments of the present specification;

FIG. 4C is a schematic diagram of an exemplary correction process according to some embodiments of the present specification;

Fig. 5A is a schematic diagram of an exemplary method for determining the difference between a target image and a reference image according to some embodiments of the present specification;

Fig. 5B is a schematic diagram of an exemplary method for determining the difference between a target image and a reference image according to some embodiments of the present specification;

Fig. 6A is a flowchart of an exemplary image processing method according to some embodiments of the present specification;

Fig. 6B is a schematic diagram of an exemplary image processing method according to some embodiments of the present specification;

Fig. 7A is a flowchart of an exemplary image processing method according to some embodiments of the present specification;

Fig. 7B is a schematic diagram of an exemplary image processing method according to some embodiments of the present specification;

8A is a flow chart of an exemplary image processing method according to some embodiments of the present specification; and

Fig. 8B is a schematic diagram of an exemplary image processing method according to some embodiments of the present specification.

Detailed ways

In order to more clearly illustrate the technical solutions of the embodiments of the present specification, the following briefly introduces the drawings that need to be used in the description of the embodiments. Apparently, the accompanying drawings in the following description are only some examples or embodiments of this specification, and those skilled in the art can also apply this specification to other similar scenarios. Unless otherwise apparent from context or otherwise indicated, like reference numerals in the figures represent like structures or operations.

It should be understood that "system", "device", "unit" and/or "module" as used herein is a method for distinguishing different components, elements, parts, parts or assemblies of different levels. However, the words may be replaced by other expressions if other words can achieve the same purpose.

As indicated in the specification and claims, the terms "a", "an", "an" and/or "the" are not specific to the singular and may include the plural unless the context clearly indicates an exception. Generally speaking, the terms "comprising" and "comprising" only suggest the inclusion of clearly identified steps and elements, and these steps and elements do not constitute an exclusive list, and the method or device may also contain other steps or elements.

The flowchart is used in this specification to illustrate the operations performed by the system according to the embodiment of this specification. It should be understood that the preceding or following operations are not necessarily performed in the exact order. Instead, various steps may be processed in reverse order or simultaneously. At the same time, other operations can be added to these procedures, or a certain step or steps can be removed from these procedures.

Fig. 1 is a schematic diagram of an application scenario of an exemplary image processing system according to some embodiments of the present specification. As shown in FIG. 1 , in some embodiments, an image processing system 100 may include an imaging device 110 , a network 120 , a terminal device 130 , a processing device 140 and a storage device 150 . Multiple components in the image processing system 100 may be connected to each other through a network 120 . For example, the imaging device 110 and the terminal device 130 may be connected or communicate through the network 120 . For another example, the imaging device 110 and the processing device 140 may be connected or communicate through the network 120 . In some embodiments, connections between components in image processing system 100 are variable. For example, the terminal device 130 may be directly connected to the processing device 140 .

The imaging device 110 may be used to scan an object in a detection area or a scanning area to obtain imaging data of the object. In some embodiments, objects may include biological objects and/or non-biological objects. For example, an object may be animate or inanimate organic and/or inorganic matter.

In some embodiments, imaging device 110 may be a non-invasive imaging device used for disease diagnosis or research purposes. For example, imaging device 110 may include a single modality scanner and/or a multimodal scanner. Single modality scanners may include, for example, ultrasound scanners, X-ray scanners, computed tomography (CT) scanners, magnetic resonance imaging (MRI) scanners, sonography, positron emission tomography (PET) scanners , optical coherence tomography (OCT) scanner, ultrasound (US) scanner, intravascular ultrasound (IVUS) scanner, near-infrared spectroscopy (NIRS) scanner, far-infrared (FIR) scanner, etc., or any combination thereof. Multimodal scanners may include, for example, X-ray imaging-magnetic resonance imaging (X-ray-MRI) scanners, positron emission tomography-X-ray imaging (PET-X-ray) scanners, single photon emission computed tomography-MRI Resonance imaging (SPECT-MRI) scanner, positron emission tomography-computed tomography (PET-CT) scanner, digital subtraction angiography-magnetic resonance imaging (DSA-MRI) scanner, etc. or any combination thereof. The scanners described above are for illustration purposes only and are not intended to limit the scope of this specification. As an example, the imaging device 110 may include an MRI scanner 111 and a CT scanner 112, wherein the MRI scanner 111 may be used to acquire MRI data of a subject to generate an MR image, and the CT scanner 112 may be used to acquire CT data of a subject , to generate a CT image.

Network 120 may include any suitable network capable of facilitating the exchange of information and/or data for image processing system 100 . In some embodiments, at least one component of the image processing system 100 (for example, the imaging device 110, the terminal device 130, the processing device 140, the storage device 150) can exchange information and /or data. For example, the processing device 140 may acquire imaging data of the object from the imaging device 110 through the network 120 . Network 120 may include public networks (e.g., the Internet), private networks (e.g., local area networks (LANs)), wired networks, wireless networks (e.g., 802.11 networks, Wi-Fi networks), frame relay networks, virtual private networks (VPN), Satellite Network, Telephone Network, Router, Hub, Switch, Fiber Optic Network, Telecom Network, Intranet, Wireless Local Area Network (WLAN), Metropolitan Area Network (MAN), Public Switched Telephone Network (PSTN), ^BluetoothTM network, ZigBee ^™ network, Near Field Communication (NFC) network, etc. or any combination thereof. In some embodiments, network 120 may include at least one network access point. For example, network 120 may include wired and/or wireless network access points, such as base stations and/or Internet exchange points, through which at least one component of image processing system 100 may connect to network 120 to exchange data and/or information.

The terminal device 130 may communicate with and/or be connected to the imaging device 110 , the processing device 140 and/or the storage device 150 . For example, the user may interact with the imaging device 110 through the terminal device 130 to control one or more components of the imaging device 110 . In some embodiments, the terminal device 130 may include a mobile device 131, a tablet computer 132, a notebook computer 133, etc. or any combination thereof. For example, mobile device 131 may include a mobile controller handle, personal digital assistant (PDA), smartphone, etc., or any combination thereof.

The processing device 140 may process data and/or information acquired from the imaging device 110 , the terminal device 130 and/or the storage device 150 . For example, the processing device 140 may acquire the imaging data of the object from the imaging device 110, and determine a target image of the object and at least one reference image, wherein the imaging devices corresponding to the target image and the reference image are different (for example, respectively corresponding to MRI scanners and CT scanners). For another example, the processing device 140 may correct the target image based on at least one reference image.

In some embodiments, processing device 140 may be a single server or a group of servers. Server groups can be centralized or distributed. In some embodiments, processing device 140 may be local or remote. For example, processing device 140 may access information and/or data from imaging device 110 , terminal device 130 and/or storage device 150 via network 120 . As another example, the processing device 140 may be directly connected to the imaging device 110, the terminal device 130 and/or the storage device 150 to access information and/or data. In some embodiments, the processing device 140 may be implemented on a cloud platform. For example, a cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an inter-cloud, a multi-cloud, etc., or any combination thereof.

In some embodiments, processing device 140 may include one or more processors (eg, single-chip processors or multi-chip processors). For example only, the processing device 140 may include a central processing unit (CPU), an application specific integrated circuit (ASIC), an application specific instruction set processor (ASIP), a graphics processing unit (GPU), a physical processing unit (PPU), a digital signal processing DSP, Field Programmable Gate Array (FPGA), Programmable Logic Device (PLD), Controller, Microcontroller Unit, Reduced Instruction Set Computer (RISC), Microprocessor, etc. or any combination thereof. In some embodiments, the processing device 140 (or all or part of its functionality) may be part of the imaging device 110 or the terminal device 130 .

Storage device 150 may store data, instructions and/or any other information. In some embodiments, the storage device 150 may store data acquired from the imaging device 110 , the terminal device 130 and/or the processing device 140 . For example, the storage device 150 may store imaging data acquired from the imaging device 110 and related information thereof. As another example, the storage device 150 may store images (eg, target images, reference images) generated based on imaging data. In some embodiments, storage device 150 may store data and/or instructions that processing device 140 executes or uses to perform the exemplary methods described in this specification. In some embodiments, the storage device 150 may include mass storage, removable storage, volatile read-write storage, read-only memory (ROM), etc., or any combination thereof. In some embodiments, the storage device 150 can be implemented on a cloud platform.

In some embodiments, the storage device 150 may be connected to the network 120 to communicate with at least one other component in the image processing system 100 (eg, the imaging device 110 , the terminal device 130 , the processing device 140 ). At least one component in the image processing system 100 can access data stored in the storage device 150 (eg, a target image of a subject, a reference image, etc.) through the network 120 . In some embodiments, storage device 150 may be part of processing device 140 .

It should be noted that the above description is provided for illustrative purposes only and is not intended to limit the scope of this specification. Those skilled in the art can make various changes and modifications under the guidance of the contents of this specification. The features, structures, methods, and other features of the exemplary embodiments described in this specification can be combined in various ways to obtain additional and/or alternative exemplary embodiments.

Fig. 2 is a block diagram of an exemplary image processing system according to some embodiments of the present specification. As shown in FIG. 2 , in some embodiments, the image processing system 200 may include a target image acquisition module 210 , a reference image acquisition module 220 and a correction module 230 . In some embodiments, the image processing system 200 may be implemented by the processing device 140 .

The target image acquisition module 210 may be used to acquire target images of objects. Target images can refer to images that have target features or meet target requirements. For more information about the acquisition of the target image, refer to step 310 in FIG. 3 and related descriptions.

The reference image acquisition module 220 can be used to acquire at least one reference image of the object. The reference image can be used to correct the target image. In some embodiments, the imaging device corresponding to the reference image may be different from the imaging device corresponding to the target image. In some embodiments, the reference image and the target image may respectively correspond to different modalities of the imaging device. For more information about reference image acquisition, refer to step 320 in FIG. 3 and related descriptions.

The correction module 230 can be used to correct the target image based on at least one reference image. In some embodiments, the correction module 230 may preprocess the target image first, and then correct the preprocessed target image based on at least one reference image. In some embodiments, the correction module 230 may determine a difference between the target image and at least one reference image, and correct the target image based on the difference. For more information about reference image acquisition, refer to step 330 in FIG. 3 and related descriptions.

It should be understood that the image processing system 200 and its modules shown in FIG. 2 can be implemented in various ways, for example, implemented by hardware, software, or a combination of software and hardware. The system and its modules in this specification can not only be realized by hardware circuits such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc. , can also be realized by software executed by various types of processors, for example, and can also be realized by a combination of the above-mentioned hardware circuits and software (for example, firmware).

It should be noted that the above description of the image processing system 200 and its modules is only for convenience of description, and does not limit this description to the scope of the illustrated embodiments. It can be understood that for those skilled in the art, after understanding the principle of the system, it is possible to combine various modules arbitrarily, or form a subsystem to connect with other modules without departing from this principle.

Fig. 3 is a flowchart of an exemplary image processing method according to some embodiments of this specification. In some embodiments, the process 300 may be executed by the processing device 140 or the image processing system 200 . For example, the process 300 may be stored in a storage device (for example, the storage device 150, the storage unit of the processing device 140) in the form of a program or an instruction, and when the processor or the module shown in FIG. 2 executes the program or the instruction, the process may be implemented. 300. In some embodiments, process 300 may be accomplished with one or more additional operations not described below, and/or without one or more operations discussed below. In addition, the order of operations shown in FIG. 3 is not limiting.

Step 310, acquiring a target image of the object. In some embodiments, this step 310 may be performed by the processing device 140 or the image processing system 200 (eg, the target image acquisition module 210 ).

In some embodiments, objects may include biological objects and/or non-biological objects. For example, an object may be animate or inanimate organic and/or inorganic matter. As another example, an object may include a specific part, organ, and/or tissue of a patient. By way of example only, an object may include a patient's brain, neck, heart, lungs, etc., or any combination thereof.

The target image can be an image whose target features meet the target requirements. For example, the target image may include an MR image with high soft tissue resolution accuracy. For another example, the target image may include a CT image with high spatial position accuracy. The target features or target requirements are related to the actual needs of images or diagnosis and treatment, and are not limited in this specification.

In some embodiments, the target image may include a two-dimensional (2D) image, a three-dimensional (3D) image, a four-dimensional (4D) image (eg, a time sequence of 3D images), etc., or any combination thereof. In some embodiments, the target images may include CT images, MR images, PET images, SPECT images, ultrasound images, X-ray images, etc. or any combination thereof.

In some embodiments, the target image acquisition module 210 may acquire imaging data of the object from an imaging device (for example, the imaging device 110 ), and determine the target image based on the imaging data. In some embodiments, the target image acquisition module 210 can also perform corrections (for example, random correction, detector normalization, scatter correction, attenuation correction, etc.) or preprocessing operations (for example, resizing, image resampling, image normalization, etc.) to determine the target image. In some embodiments, the target image acquisition module 210 may determine the target image based on the imaging data through an image reconstruction algorithm. Taking MR images as an example, exemplary MR image reconstruction algorithms may include Fourier transform algorithms, back-projection algorithms (for example, convolution back-projection algorithms or filter back-projection algorithms), iterative reconstruction algorithms, and the like. In some embodiments, the target image acquisition module 210 can acquire pre-stored target images from a storage device (eg, the storage device 150 ) or an external storage device (eg, a medical image database).

Step 320, acquiring at least one reference image of the object. In some embodiments, this step 320 may be performed by the processing device 140 or the image processing system 200 (eg, the reference image acquisition module 220 ).

The reference image can be used to correct the target image. In some embodiments, features or characteristics of the reference image are at least partially different from corresponding features or characteristics of the target image. In some embodiments, a feature or characteristic of the reference image is at least partially superior to a corresponding feature or characteristic of the target image. In some embodiments, the reference image is complementary in some dimension or aspect to features or characteristics of the target image. For example, a target image has feature A and feature B, wherein feature A (ie, the target feature) has higher precision, and feature B has lower precision. Correspondingly, the reference image may be an image with higher precision of the feature B. For another example, the target image has feature A, feature B, feature C, and feature D, wherein feature A (namely, the target feature) has higher accuracy, while feature B, feature C, and feature D have lower accuracy. Correspondingly, the reference image can be an image with high accuracy of feature B, feature C, and feature D; Higher precision. Specifically, for example, the target image may include a magnetic resonance imaging (MRI) image with high soft tissue resolution accuracy and low spatial position accuracy, and the reference image may include a computed tomography (CT) image with high spatial position accuracy. , scanned images of radar sensors, X-ray images, etc. or any combination thereof.

In some embodiments, the acquisition manner of the reference image is similar to the acquisition manner of the target image, which will not be repeated here.

In some embodiments, the imaging device corresponding to the reference image may be different from the imaging device corresponding to the target image. For example, the imaging device corresponding to the target image may be an MRI device, and the imaging device corresponding to the reference image may be an ultrasound device, X-ray device, CT device, PET device, OCT device, IVUS device, NIRS device, FIR device, etc. or any combination. In some embodiments, the reference image and the target image may respectively correspond to different modalities of the imaging device. For example, the imaging device may be an MRI-CT device, the target image may correspond to the MRI modality of the imaging device, and the reference image may correspond to the CT modality of the imaging device.

In some embodiments, at least one reference image may include multiple reference images. In some embodiments, multiple reference images may be acquired by different imaging devices. Correspondingly, multiple reference images may have features of different dimensions or aspects. For example, the multiple reference images may include a first reference image and a second reference image, wherein the first reference image may be acquired by a CT device, and the second reference image may be acquired by a PET device. In some embodiments, multiple reference images may be acquired by the same imaging device. In some embodiments, multiple reference images may be collected by the same imaging device based on different imaging conditions (eg, imaging angle, sampling frequency, ray intensity, magnetic field intensity, etc.). For example, multiple reference images may all be collected by CT equipment, but their corresponding imaging angles are different.

Through different imaging devices or different imaging conditions, reference images corresponding to different dimensions or different aspects can be collected, and correspondingly, target images can be corrected from different dimensions or different aspects to improve the multi-dimensional correction effect.

Step 330, based on at least one reference image, correct the target image. In some embodiments, this step 330 may be performed by the processing device 140 or the image processing system 200 (eg, the correction module 230 ).

In some embodiments, the processing device 140 may preprocess the target image first, and then correct the preprocessed target image based on at least one reference image.

By preprocessing the target image, some factors that may affect the image quality can be reduced or eliminated in advance, thereby improving the subsequent correction effect. In some embodiments, taking the target image as an example of an MRI image, the influence of inhomogeneity of the main magnetic field (B0 field) may be reduced or eliminated through preprocessing. For example, the processing device 140 may preprocess the target image through a compensation algorithm. Exemplary compensation algorithms may include 3D phase unwarping methods (eg, path-guided methods, least-p-norm methods, etc.), 2D phase unwarping methods, and the like. For another example, the processing device 140 may obtain a B0 field correction matrix in advance, and preprocess the target image by using the B0 field correction matrix.

Generally speaking, for different objects or under different imaging parameters or sequence parameters, the impact of B0 field inhomogeneity is different. Therefore, reducing or eliminating the influence of the B0 field in advance through preprocessing can improve the correction effect of subsequent images.

In some embodiments, preprocessing may also include adjusting resolution, adjusting contrast, adjusting signal-to-noise ratio, spatial domain processing (eg, smoothing, edge detection, sharpening, etc.), etc., or any combination thereof.

In some embodiments, processing device 140 may determine a difference between the target image and at least one reference image, and correct the target image based on the difference.

In some embodiments, the difference may reflect the difference between the target image and at least one reference image in various dimensions or aspects (eg, spatial position, gray value, gradient value, resolution, brightness, etc.). In some embodiments, the difference may reflect the difference between the target image and the at least one reference image in angles of features other than target features. For example, a target image has a feature A and a feature B, wherein feature A has a higher accuracy and feature B has a lower accuracy. The reference image may be an image with higher accuracy of the feature B. Accordingly, the difference may be the difference between the feature B of the target image and the feature B of the reference image. In some embodiments, the difference can be represented by numerical values, vectors, matrices, models, images, and the like.

In some embodiments, the difference may include a spatial position difference of the same location point of the object in the target image and at least one reference image. For example, the processing device 140 may map the target image and the at least one reference image to the same spatial coordinate system, and determine the difference in the spatial coordinates of the same position point of the object in the target image and the at least one reference image (for example, between the spatial coordinates difference).

As an example only, FIG. 4A is a schematic diagram of an exemplary reference image shown according to some embodiments of this specification, and FIG. 4B is a schematic diagram of an exemplary target image shown according to some embodiments of this specification. The dots in FIG. The dots in 4B are in one-to-one correspondence, the dots corresponding to each other correspond to the same position of the object, and the difference in coordinates between the dots corresponding to each other is the difference between the target image and the reference image.

In some embodiments, the processing device 140 can register the at least one reference image and the target image, and determine the at least one reference image and the target image based on the first registration result of the at least one reference image and the target image. difference. For example, after registering the at least one reference image and the target image, the processing device 140 may directly perform subtraction processing on the registered at least one reference image and the target image, so as to determine the difference between the two. For another example, after registering at least one reference image and the target image, the processing device 140 may directly perform division processing on the registered at least one reference image and the target image, so as to determine the difference between the two. In some embodiments, exemplary image registration algorithms may include mean absolute difference (MAD), sum of absolute error (SAD), sum of squared error (SSD), sum of squared error (MSD), normalization Product correlation algorithm (NCC), sequential similarity detection algorithm (SSDA), local gray value coding algorithm, etc. or any combination thereof.

In some embodiments, the processing device 140 may input the at least one reference image and the target image into the first model, and based on the output of the first model, determine the difference of the at least one reference image and the target image. For more information about the first model, please refer to FIG. 5A and its related description.

In some embodiments, for each of the plurality of location points of the object, the processing device 140 may determine the point difference (e.g., spatial location difference, gray value difference) of the location point in the target image and at least one reference image. , gradient value difference, resolution difference, brightness difference, etc.), and based on the multiple point differences corresponding to the multiple position points, a difference model is constructed, and the difference model can reflect the difference between at least one reference image and the target image. See Figure 5B and its associated description for more on the difference-based model.

In some embodiments, the processing device 140 may adjust the corresponding features of the target image by difference to correct the target image. For example, in combination with the above, the target image has feature A and feature B, wherein feature A has higher precision, and feature B has lower precision. The reference image may be an image with higher accuracy of the feature B. The difference may be the difference between feature B of the target image and feature B of the reference image. Correspondingly, the processing device 140 can adjust the feature B of the target image based on the difference so that it is close to the feature B of the reference image. As an example only, FIG. 4C is a schematic diagram of an exemplary correction process according to some embodiments of the present specification. The arrows in FIG. 4C represent the coordinate differences of multiple position points of the object between at least one reference image and the target image, Correspondingly, the processing device 140 can adjust the coordinate values of each point in the target image based on the coordinate difference, thereby improving the accuracy of its spatial position.

In some embodiments, the processing device 140 may input the difference and the target image into the machine learning model, and based on the output of the machine learning model, determine the corrected target image.

In some embodiments, at least one reference image may include multiple reference images. Processing device 140 may determine differences between the target image and each of the plurality of reference images. The processing device 140 may perform comprehensive processing on the differences respectively corresponding to the multiple reference images. For example, the processing device 140 may perform weighting processing on differences respectively corresponding to multiple reference images. Further, the processing device 140 may correct the target image based on the comprehensive processing result. For more information about comprehensively processing the differences corresponding to multiple reference images, please refer to FIG. 6A, FIG. 6B and their related descriptions.

In some embodiments, processing device 140 may determine differences between the target image and each of the plurality of reference images. The processing device 140 may correct the target image to determine an intermediate corrected image based on the difference. Further, the processing device 140 may determine the corrected target image based on the multiple intermediate corrected images. For more information on determining the intermediate corrected image, refer to FIG. 7A, FIG. 7B and their related descriptions.

In some embodiments, the processing device 140 can register the at least one reference image and the target image, and determine the corrected target image based on the second registration result of the at least one reference image and the target image. For example, after registering the at least one reference image and the target image, the processing device 140 may directly perform subtraction processing on the registered at least one reference image and the target image, so as to determine the corrected target image. For another example, after registering the at least one reference image and the target image, the processing device 140 may directly perform division processing on the registered at least one reference image and the target image, so as to determine the corrected target image.

In some embodiments, the processing device 140 may input at least one reference image and the target image into the second model. And based on the output of the second model, the corrected target image is determined. For more information about the second model, refer to FIG. 8A, FIG. 8B and their related descriptions.

According to an embodiment of the present specification, the target image may be corrected based on a difference between at least one reference image of the object and the target image, wherein features or characteristics of the reference image are at least partially better than corresponding features or characteristics of the target image. Correspondingly, while retaining the target features of the target image, other features of the target image can be adjusted to be close to the corresponding features of the reference image, thereby enriching the features of the target image, improving the image quality of the target image, and providing comprehensive image information to improve subsequent possible therapeutic effects.

It should be noted that the above description about the process 300 is only for illustration and description, and does not limit the scope of application of this description. For those skilled in the art, various modifications and changes can be made to the process 300 under the guidance of this description. However, such modifications and changes are still within the scope of this specification.

Fig. 5A is a schematic diagram of an exemplary method for determining the difference between a target image and a reference image according to some embodiments of the present specification.

As shown in FIG. 5A, in some embodiments, a target image 510 and at least one reference image 515 may be input into a first model 520, and based on the output of the first model 520, the difference between the target image and at least one reference image may be determined 530.

In some embodiments, the first model 520 may be a convolutional neural network model (Convolutional Neural Network, CNN), a deep neural network model (Deep Neural Network, DNN), a recurrent neural network model (Recurrent Neural Network, RNN), a graph Neural network model (Graph Neural Network, GNN), Generative Adversarial Network model (Generative Adversarial Network, GAN), etc. or any combination thereof.

In some embodiments, the first model 520 may be determined based on multiple sets of first training samples 540 through training. Each set of first training samples 540 may include a sample target image 541 of the sample object, at least one sample reference image 542 of the sample object, and a corresponding sample difference 543, wherein the sample target image 541 and at least one sample reference image 542 are training data, and the corresponding sample difference 543 is a label.

In some embodiments, the sample target image 541 and at least one sample reference image 542 correspond to different imaging devices. In some embodiments, at least one sample reference image 542 corresponds to a different imaging device. The relationship between the sample target image 541 and the sample reference image 542 is similar to the relationship between the target image and the reference image, and a more specific description can be found in FIG. 3 .

In some embodiments, the sample difference 543 between the sample target image 541 and at least one sample reference image 542 can be marked manually by a user, or automatically marked by the image processing system 100 .

In some embodiments, the processing device 140 (or other processing device) takes the sample target image 541 and at least one sample reference image 542 as input, and uses the corresponding sample difference 543 as supervision to train the first model 520, by machine The learning algorithm (for example, stochastic gradient descent method) updates the parameters of the first model 520 to minimize the first loss function until the model training is completed; or the training is stopped after the number of iteration training reaches a certain number of times.

In some embodiments, the first loss function may be a perceptual loss function. In some embodiments, the first loss function may also be other loss functions, for example, a square loss function, a logistic regression loss function, and the like.

Fig. 5B is a schematic diagram of an exemplary method for determining the difference between a target image and a reference image according to some embodiments of the present specification.

As shown in FIG. 5B, in some embodiments, for each of the plurality of location points of the object, the processing device 140 may determine the point difference 570 of the location point in the target image 560 and at least one reference image 565, and Based on the multiple point differences corresponding to the multiple position points, the difference 590 between the target image 560 and at least one reference image 565 is determined by constructing a difference model 580 .

In some embodiments, point differences may include spatial position differences, gray value differences, gradient value differences, resolution differences, brightness differences, etc. or any combination thereof. For example, for the spatial position difference, the processing device 140 may map the target image and at least one reference image to the same spatial coordinate system, and determine the spatial coordinate difference ( For example, the difference between spatial coordinates). Specifically, for example, taking a specific position point of an object as an example, the coordinates of the position point in the target image 560 are (x _a , y _a , z _a ), and the coordinates of the position point in at least one reference image 565 are (x _b , y _b , z _b ), then the spatial position difference d of the position point in the target image 560 and at least one reference image 565 can be determined based on the following formula (1) or formula (2):

d=((x _a -x _b ),(y _a -y _b ),(z _a -z _b )); (1)

d＝((x _b -x _a ),(y _b -y _a ),(y _b -y _a )) (2)

In some embodiments, the processing device 140 may construct a difference model 580 based on a plurality of point differences respectively corresponding to a plurality of position points, so as to determine a difference 590 between at least one reference image and the target image. In some embodiments, the processing device 140 may construct the difference model 590 through a data modeling method. In some embodiments, the processing device 140 may construct the difference model 590 through a three-dimensional modeling method. In some embodiments, the processing device 140 can construct the difference model 590 by principal component analysis, simulation, regression analysis, cluster analysis and other methods. In some embodiments, the difference 590 may be represented by numerical values, vectors, matrices, models, images, and the like.

Taking the spatial position difference as an example, after determining the corresponding spatial position differences (for example, spatial coordinate differences) among the multiple position points, the processing device 140 can determine the spatial position corresponding to the intermediate position point between adjacent position points by interpolation. difference, so as to determine the spatial position difference corresponding to a large number of position points. The processing device 140 can build a difference model 580 (for example, a vector diagram as shown in FIG. 4C ) based on a large number of spatial position differences, and the difference model 580 can reflect the spatial position difference between at least one reference image and the target image as a whole.

Fig. 6A is a flowchart of an exemplary image processing method according to some embodiments of the present specification. In some embodiments, the process 600 may be executed by the processing device 140 or the image processing system 200 . For example, the process 600 may be stored in a storage device (for example, the storage device 150, the storage unit of the processing device 140) in the form of a program or an instruction, and when the processor or the module shown in FIG. 2 executes the program or the instruction, the process may be implemented. 600. In some embodiments, process 600 may be accomplished with one or more additional operations not described below, and/or without one or more operations discussed below. In addition, the order of operations shown in FIG. 6A is not limiting. In some embodiments, correcting the target image described in operation 330 in FIG. 3 may be performed according to process 600 .

Step 610, determine the difference between the target image and each of the plurality of reference images (difference 660, difference 665, etc. as shown in FIG. 6B). In some embodiments, this step 610 may be performed by the processing device 140 or the image processing system 200 (eg, the correction module 230 ).

The difference may reflect the difference between the target image and each of the multiple reference images in various dimensions or aspects (for example, spatial position, gray value, gradient value, resolution, brightness, etc.). In some embodiments, the processing device 140 may determine the difference between the target image and each of the plurality of reference images through the first model or the difference model. For more information on determining the difference, refer to FIG. 5A, FIG. 5B and their related descriptions.

Step 620, perform comprehensive processing on the differences corresponding to the multiple reference images. In some embodiments, this step 620 may be performed by the processing device 140 or the image processing system 200 (eg, the correction module 230 ).

In some embodiments, the processing device 140 may perform weighting processing on the differences respectively corresponding to the multiple reference images. For example, the processing device 140 may assign different weights to the differences corresponding to the multiple reference images, and determine the overall difference corresponding to the multiple reference images. In some embodiments, the weight of the difference corresponding to each reference image may be determined by the user or determined by the image processing system 100 according to image processing requirements. For example, the target image has feature A, feature B, feature C, and feature D, wherein, feature A (namely, the target feature) has higher accuracy, while feature B, feature C, and feature D have lower accuracy. The reference image may include three, feature B, feature C, and feature D with relatively high precision. Correspondingly, the difference between each reference image and the target image may include a feature B difference, a feature C difference, and a feature C difference. Assuming that the final image processing goal is to correct feature B first, then feature C, and then feature D, correspondingly, the weights of the differences corresponding to the three reference images are sequentially reduced.

As an example only, the processing device 140 may determine the overall difference corresponding to multiple reference images based on the following formula (3):

Among them, d _t represents the overall difference corresponding to multiple reference images, r _i represents the weight of the difference corresponding to the i-th reference image, d _i represents the difference corresponding to the i-th reference image, n represents the total number of multiple reference images, n is a positive integer.

In some embodiments, the processing device 140 may perform average processing or weighted average processing on differences respectively corresponding to multiple reference images. For example, multiple reference images may correspond to the same imaging device or to the same imaging condition, but due to inevitable systematic errors, certain differences or deviations inevitably exist among the multiple reference images. Correspondingly, by performing averaging processing or weighted average processing on the corresponding differences, the errors can be balanced, thereby improving the subsequent correction effect.

Step 630, based on the comprehensive processing result (such as the comprehensive processing result 670 shown in FIG. 6B), determine the corrected target image (the corrected target image 680 shown in FIG. 6B). In some embodiments, this step 630 may be performed by the processing device 140 or the image processing system 200 (eg, the correction module 230 ).

In some embodiments, the processing device 140 can comprehensively adjust the corresponding features of the target image to make it close to the corresponding features of the reference image by integrating the processing results, so as to correct the target image. In some embodiments, the integrated processing results may be represented in the form of numerical values, vectors, matrices, models, images, and the like. For example, the processing device 140 may adjust pixel values or related information of each pixel of the target image based on the matrix corresponding to the comprehensive processing result, so as to correct the target image.

By comprehensively processing the differences corresponding to multiple reference images, not only can different weights of different features to be corrected be considered, but also possible system errors can be balanced, thereby improving the image correction effect.

Fig. 7A is a flowchart of an exemplary image processing method according to some embodiments of the present specification. In some embodiments, the process 700 may be executed by the processing device 140 or the image processing system 200 . For example, the process 700 may be stored in a storage device (for example, storage device 150, a storage unit of the processing device 140) in the form of a program or an instruction, and when the processor or the module shown in FIG. 2 executes the program or the instruction, the process may be implemented. 700. In some embodiments, process 700 may be accomplished with one or more additional operations not described below, and/or without one or more operations discussed below. In addition, the order of operations shown in FIG. 7A is not limiting. In some embodiments, correcting the target image described in operation 330 in FIG. 3 may be performed according to process 700 .

Step 710, determining the difference between the target image and each of the plurality of reference images. In some embodiments, this step 710 may be performed by the processing device 140 or the image processing system 200 (eg, the correction module 230 ).

For example, as shown in FIG. 7B , processing device 140 may determine difference 770 between target image 750 and reference image 760 , difference 775 between target image 755 and reference image 765 , and so on. For more information on determining the difference, refer to FIG. 5A, FIG. 5B and their related descriptions.

Step 720, based on the difference, correct the target image to determine an intermediate corrected image. In some embodiments, this step 720 may be performed by the processing device 140 or the image processing system 200 (eg, the correction module 230 ).

For example, as shown in FIG. 7B , processing device 140 may correct target image 750 based on difference 770 to determine intermediate corrected image 780; processing device 140 may correct target image 755 based on difference 775 to determine intermediate corrected image 785;

In some embodiments, the processing device 140 can adjust the corresponding features of the target image to be close to the corresponding features of the reference image through the difference, so as to correct the target image. Step 330 or step 630 can be seen for correcting the content of the target image based on the difference.

Step 730: Determine a corrected target image based on the multiple intermediate corrected images. In some embodiments, this step 730 may be performed by the processing device 140 or the image processing system 200 (eg, the correction module 230 ).

For example, processing device 140 may determine final corrected target image 790 based on intermediate corrected image 780, intermediate corrected image 785, and the like.

In some embodiments, the processing device 140 may perform weighting processing on multiple intermediate corrected images. For example, the processing device 140 may assign different weights to the multiple intermediate corrected images, and perform weighting processing based on the weights. In some embodiments, the weight of each intermediate corrected image can be determined by the user or determined by the image processing system 100 according to image processing requirements. For example, the target image has feature A, feature B, feature C, and feature D, wherein, feature A (namely, the target feature) has higher accuracy, while feature B, feature C, and feature D have lower accuracy. The reference image may include three, feature B, feature C, and feature D with relatively high precision. Correspondingly, the target image is corrected based on the difference between each reference image and the target image, and an intermediate corrected image B, an intermediate corrected image C, and an intermediate corrected image D are respectively obtained. Assuming that the final image processing goal is to correct feature B first, then feature C, and then feature D, correspondingly, the weights of the differences corresponding to the three intermediate corrected images are sequentially reduced.

As an example only, the processing device 140 may determine the final corrected target image based on the following formula (4):

Among them, I _t represents the final corrected target image, W _i represents the weight corresponding to the i-th intermediate corrected image, R _i represents the i-th intermediate corrected image, and n represents the total number of multiple reference images (or intermediate corrected images) , n is a positive integer.

In some embodiments, the processing device 140 may perform average processing or weighted average processing on the multiple intermediate corrected images respectively corresponding to the multiple reference images. For example, multiple reference images may correspond to the same imaging device or to the same imaging conditions, but due to unavoidable systematic errors, certain differences or deviations inevitably exist among the multiple intermediate corrected images. Correspondingly, by performing averaging processing or weighted average processing on the corresponding differences, the errors can be balanced, thereby improving the subsequent correction effect.

By performing correction based on multiple reference images respectively, and performing subsequent comprehensive processing on multiple intermediate corrected images to determine the final corrected target image, the calculation amount of a single correction can be reduced, and the different weights of different features to be corrected can be considered. Possible systematic errors can be equalized, thereby improving the image correction effect.

Fig. 8A is a flowchart of an exemplary image processing method according to some embodiments of the present specification. In some embodiments, the process 800 may be executed by the processing device 140 or the image processing system 200 . For example, the process 800 may be stored in a storage device (for example, the storage device 150, the storage unit of the processing device 140) in the form of a program or an instruction, and when the processor or the module shown in FIG. 2 executes the program or the instruction, the process may be implemented. 800. In some embodiments, process 800 may be accomplished with one or more additional operations not described below, and/or without one or more operations discussed below. In addition, the order of operations as shown in FIG. 8A is not limiting. In some embodiments, correcting the target image described in operation 330 in FIG. 3 may be performed according to process 800 .

Step 810, input at least one reference image and target image into the second model. In some embodiments, this step 810 may be performed by the processing device 140 or the image processing system 200 (eg, the correction module 230 ). For more information on the second model, refer to FIG. 8B and its related description.

Step 820, based on the output of the second model, determine the corrected target image. In some embodiments, this step 820 may be performed by the processing device 140 or the image processing system 200 (eg, the correction module 230 ).

As shown in FIG. 8B , in some embodiments, the target image 860 and at least one reference image 865 may be input into the second model 870 to obtain a corrected target image 880 .

In some embodiments, the first model 870 may be a convolutional neural network model (Convolutional Neural Network, CNN), a deep neural network model (Deep Neural Network, DNN), a recurrent neural network model (Recurrent Neural Network, RNN), a graph Neural network model (Graph Neural Network, GNN), Generative Adversarial Network model (Generative Adversarial Network, GAN), etc. or any combination thereof.

In some embodiments, the second model 870 may be determined based on multiple sets of second training samples 890 for training. Each set of second training samples 890 may include a sample target image 891 of the sample object, at least one sample reference image 892 of the sample object and a corresponding sample corrected image 893, wherein the sample target image 891 and at least one sample reference image 892 are For the training data, the corresponding sample corrected image 893 is a label.

In some embodiments, the sample target image 891 and the at least one sample reference image 892 correspond to different imaging devices. In some embodiments, at least one sample reference image 892 corresponds to a different imaging device. The relationship between the sample target image 891 and the sample reference image 892 is similar to the relationship between the target image and the reference image, and a more specific description can be found in FIG. 3 .

In some embodiments, the corresponding sample corrected image 893 between the sample target image 891 and at least one sample reference image 892 can be manually marked by a user (eg, manually modified or edited by a doctor), or can be automatically marked by the image processing system 100 .

In some embodiments, the processing device 140 (or other processing device) takes the sample target image 891 and at least one sample reference image 892 as input, and uses the corresponding sample corrected image 893 as supervision to train the second model 870 by The learning algorithm (for example, stochastic gradient descent method) updates the parameters of the second model 870 to minimize the second loss function until the model training is completed; or the training is stopped after the number of iteration training reaches a certain number of times.

In some embodiments, the second loss function may be a perceptual loss function. In some embodiments, the second loss function may also be other loss functions, for example, a square loss function, a logistic regression loss function, and the like.

In some embodiments of this specification, (1) the target image and the reference image are acquired through different imaging methods, and the target image is corrected based on the difference between at least one reference image and the target image, which can be adjusted while retaining the target features of the target image. Other features of the target image are close to the corresponding features of the reference image, thereby enriching the features of the target image and improving the image quality of the target image; (2) determining the difference between the reference image and the target image by using a machine learning model or constructing a difference model, which can Improve the accuracy, efficiency and comprehensiveness of determining the difference; (3) through the comprehensive processing of multiple reference images, different image processing requirements can be met, and the image correction effect can be improved.

Some embodiments of this specification also provide an image processing device, which includes: at least one storage medium storing computer instructions; at least one processor executing the computer instructions to implement the image processing method described in this specification. For many technical details, refer to the relevant descriptions in FIG. 1 to FIG. 8B , and details are not repeated here.

Some embodiments of this specification also provide a computer-readable storage medium, which stores computer instructions. When a computer reads the computer instructions, the computer executes the image processing method described in this specification. For more technical details, please refer to The relevant descriptions of FIG. 1 to FIG. 8B are not repeated here.

The basic concept has been described above, obviously, for those skilled in the art, the above detailed disclosure is only an example, and does not constitute a limitation to this description. Although not expressly stated here, those skilled in the art may make various modifications, improvements and corrections to this description. Such modifications, improvements and corrections are suggested in this specification, so such modifications, improvements and corrections still belong to the spirit and scope of the exemplary embodiments of this specification.

Meanwhile, this specification uses specific words to describe the embodiments of this specification. For example, "one embodiment", "an embodiment", and/or "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of this specification. Therefore, it should be emphasized and noted that two or more references to "an embodiment" or "an embodiment" or "an alternative embodiment" in different places in this specification do not necessarily refer to the same embodiment . In addition, certain features, structures or characteristics in one or more embodiments of this specification may be properly combined.

In addition, unless explicitly stated in the claims, the order of processing elements and sequences described in this specification, the use of numbers and letters, or the use of other names are not used to limit the sequence of processes and methods in this specification. While the foregoing disclosure has discussed by way of various examples some embodiments of the invention that are presently believed to be useful, it should be understood that such detail is for illustrative purposes only and that the appended claims are not limited to the disclosed embodiments, but rather, the claims The claims are intended to cover all modifications and equivalent combinations that fall within the spirit and scope of the embodiments of this specification. For example, although the system components described above may be implemented by hardware devices, they may also be implemented by a software-only solution, such as installing the described system on an existing server or mobile device.

In the same way, it should be noted that in order to simplify the expression disclosed in this specification and help the understanding of one or more embodiments of the invention, in the foregoing description of the embodiments of this specification, sometimes multiple features are combined into one embodiment, drawings or descriptions thereof. This method of disclosure does not, however, imply that the subject matter of the specification requires more features than are recited in the claims. Indeed, embodiment features are less than all features of a single foregoing disclosed embodiment.

In some embodiments, numbers describing the quantity of components and attributes are used. It should be understood that such numbers used in the description of the embodiments use the modifiers "about", "approximately" or "substantially" in some examples. grooming. Unless otherwise stated, "about", "approximately" or "substantially" indicates that the stated figure allows for a variation of ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximations that can vary depending upon the desired characteristics of individual embodiments. In some embodiments, numerical parameters should take into account the specified significant digits and adopt the general digit reservation method. Although the numerical ranges and parameters used in some embodiments of this specification to confirm the breadth of the range are approximations, in specific embodiments, such numerical values are set as precisely as practicable.

Each patent, patent application, patent application publication, and other material, such as article, book, specification, publication, document, etc., cited in this specification is hereby incorporated by reference in its entirety. Application history documents that are inconsistent with or conflict with the content of this specification are excluded, and documents (currently or later appended to this specification) that limit the broadest scope of the claims of this specification are also excluded. It should be noted that if there is any inconsistency or conflict between the descriptions, definitions, and/or terms used in the accompanying materials of this manual and the contents of this manual, the descriptions, definitions and/or terms used in this manual shall prevail .

Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other modifications are also possible within the scope of this description. Therefore, by way of example and not limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

An image processing system comprising:

storage devices, which store computer instructions;

A processor, connected to the storage device, when executing the computer instructions, the processor causes the system to perform the following operations:

Get the target image of the object;

acquiring at least one reference image of the object, the at least one reference image corresponding to an imaging device different from the imaging device corresponding to the target image; and

The target image is corrected based on the at least one reference image.
The system of claim 1, wherein said correcting said target image based on said at least one reference image comprises:

preprocessing the target image; and

Correcting the preprocessed target image based on the at least one reference image.
The system of claim 1, wherein said correcting said target image based on said at least one reference image comprises:

determining a difference between the target image and the at least one reference image; and

Based on the difference, the target image is corrected.
The system according to claim 3, wherein the difference comprises a spatial position difference of the same position point of the object in the target image and the at least one reference image.
The system according to claim 3, wherein said determining the difference between said target image and said at least one reference image comprises:

registering the at least one reference image and the target image; and

The difference is determined based on a first registration result of the at least one reference image and the target image.
The system according to claim 3, wherein said determining the difference between said target image and said at least one reference image comprises:

inputting said at least one reference image and said target image into a trained machine learning model; and

Based on the output of the machine learning model, the difference is determined.
The system according to claim 3, wherein said determining the difference between said target image and said at least one reference image comprises:

For each of a plurality of location points of the object, determining a point difference of the location point in the target image and the at least one reference image; and

Based on the multiple point differences respectively corresponding to the multiple position points, the difference is determined by constructing a difference model.
The system of claim 1, wherein the at least one reference image is acquired by a different imaging device.
The system of claim 1, wherein,

the at least one reference image includes a plurality of reference images; and

The correcting the target image based on the at least one reference image includes:

determining differences between the target image and each of the plurality of reference images;

performing comprehensive processing on the differences respectively corresponding to the plurality of reference images; and

Based on the comprehensive processing result, the target image is corrected.
The system of claim 1, wherein,

the at least one reference image includes a plurality of reference images; and

The correcting the target image based on the at least one reference image includes:

determining differences between the target image and each of the plurality of reference images;

based on the difference, correcting the target image to determine an intermediate corrected image; and

Based on the plurality of intermediate corrected images, a corrected target image is determined.
The system of claim 1, wherein said correcting said target image based on said at least one reference image comprises:

registering the at least one reference image and the target image; and

Based on the second registration result of the at least one reference image and the target image, a corrected target image is determined.
The system of claim 1, wherein said correcting said target image based on said at least one reference image comprises:

inputting said at least one reference image and said target image into a trained machine learning model; and

Based on the output of the machine learning model, a corrected target image is determined.
The system of claim 1, wherein,

The target image comprises a magnetic resonance (magnetic resonance imaging, MRI) image; and

The at least one reference image includes a computed tomography (CT) image.
An image processing method, comprising:

Get the target image of the object;

acquiring at least one reference image of the object, the at least one reference image corresponding to an imaging device different from the imaging device corresponding to the target image; and

The target image is corrected based on the at least one reference image.
The method according to claim 14, wherein said correcting said target image based on said at least one reference image comprises:

preprocessing the target image; and

Correcting the preprocessed target image based on the at least one reference image.
The method according to claim 14, wherein said correcting said target image based on said at least one reference image comprises:

determining a difference between the target image and the at least one reference image; and

Based on the difference, the target image is corrected.
The method according to claim 16, wherein the difference comprises a spatial position difference of the same position point of the object in the target image and the at least one reference image.
The method according to claim 16, wherein said determining the difference between said target image and said at least one reference image comprises:

registering the at least one reference image and the target image; and

The difference is determined based on a first registration result of the at least one reference image and the target image.
The method of claim 16, wherein said determining the difference between said target image and said at least one reference image comprises:

inputting said at least one reference image and said target image into a trained machine learning model; and

Based on the output of the machine learning model, the difference is determined.
The method according to claim 16, wherein said determining the difference between said target image and said at least one reference image comprises:

For each of a plurality of location points of the object, determining a point difference of the location point in the target image and the at least one reference image; and

Based on the multiple point differences respectively corresponding to the multiple position points, the difference is determined by constructing a difference model.
The method of claim 14, wherein the at least one reference image is acquired by a different imaging device.
The method of claim 14, wherein,

the at least one reference image includes a plurality of reference images; and

The correcting the target image based on the at least one reference image includes:

determining differences between the target image and each of the plurality of reference images;

performing comprehensive processing on the differences respectively corresponding to the plurality of reference images; and

Based on the comprehensive processing result, the target image is corrected.
The method of claim 14, wherein,

the at least one reference image includes a plurality of reference images; and

The correcting the target image based on the at least one reference image includes:

determining differences between the target image and each of the plurality of reference images;

based on the difference, correcting the target image to determine an intermediate corrected image; and

Based on the plurality of intermediate corrected images, a corrected target image is determined.
The method according to claim 14, wherein said correcting said target image based on said at least one reference image comprises:

registering the at least one reference image and the target image; and

Based on the second registration result of the at least one reference image and the target image, a corrected target image is determined.
The method according to claim 14, wherein said correcting said target image based on said at least one reference image comprises:

inputting said at least one reference image and said target image into a trained machine learning model; and

Based on the output of the machine learning model, a corrected target image is determined.
The method of claim 14, wherein,

The target image comprises a magnetic resonance (magnetic resonance imaging, MRI) image; and

The at least one reference image includes a computed tomography (CT) image.
A computer-readable storage medium, the storage medium stores computer instructions, and when a computer reads the computer instructions, the computer executes an image processing method, including:

Get the target image of the object;

acquiring at least one reference image of the object, the at least one reference image corresponding to an imaging device different from the imaging device corresponding to the target image; and

The target image is corrected based on the at least one reference image.
An image processing system comprising:

A target image acquisition module, configured to acquire a target image of an object;

A reference image acquisition module, configured to acquire at least one reference image of the object, the imaging device corresponding to the at least one reference image is different from the imaging device corresponding to the target image; and

A correction module, configured to correct the target image based on the at least one reference image.