Methods, systems and storage media for generating image processing models

Abstract

A method for generating image processing models is provided. The method is implemented on a computing device having at least one processor and at least one memory device. The method includes obtaining medical sample images corresponding to a plurality of tracer types; clustering the plurality of tracer types into a plurality of tracer clusters based on the medical sample images, wherein each tracer cluster includes one or more tracer types of the plurality of tracer types; for each tracer cluster, generating an image processing model corresponding to the tracer cluster by training an initial image processing model using first medical sample images of the medical sample images, wherein the first medical sample images correspond to the one or more tracer types in the tracer cluster.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to Chinese Patent Application No. 202410424881.3, filed on April 9, 2024, the entire contents of which are hereby incorporated by reference. TECHNICAL FIELD

[0002] This disclosure relates to the field of medical imaging and, in particular, to methods, systems, and storage media for generating image processing models. BACKGROUND

[0003] A tracer is a detectable and trackable marker that is often injected into a subject during a medical scan (e.g., a positron emission tomography (PET) scan) to obtain relevant information (e.g., biological metabolic information) about the subject. Medical images corresponding to different tracer types are typically processed and analyzed by different image processing models. However, the diversity of tracer types poses a challenge for training samples of the image processing models to cover all known tracer types, thus limiting the applicability of the image processing models due to the limitations of the tracer types.In addition, when processing a medical target image, it is necessary to rely on human input of tracer type information by a doctor or technician to determine the corresponding image processing model, which is prone to deterioration of the processing effect due to input errors.

[0004] Therefore, the present disclosure provides systems and methods for generating image processing models that can improve the applicability and processing accuracy of the image processing model. SUMMARY

[0005] According to one aspect of the present disclosure, a method for generating image processing models is provided. The method may be implemented on a computing device having at least one processor and at least one memory device. The method may include: obtaining medical pattern images corresponding to a plurality of tracer types; determining a plurality of tracer clusters based on the medical pattern images, each tracer cluster including one or more tracer types of the plurality of tracer types; and, for each tracer cluster, generating an image processing model corresponding to the tracer cluster by training an initial image processing model using first medical pattern images from the medical pattern images, the first medical pattern images corresponding to the one or more tracer types in the tracer cluster.

[0006] According to another aspect of the present disclosure, a system is provided. The system may include at least one storage medium storing a set of instructions and at least one processor configured to communicate with the at least one storage medium.In executing the set of instructions, the at least one processor may be directed to cause the system to perform operations including: obtaining medical sample images corresponding to a plurality of tracer types; determining the plurality of tracer types into a plurality of tracer clusters based on the medical sample images, wherein each tracer cluster includes one or more tracer types of the plurality of tracer types; and for each tracer cluster, generating an image processing model corresponding to the tracer cluster by training an initial image processing model using first medical sample images from the medical sample images, wherein the first medical sample images correspond to the one or more tracer types in the tracer cluster.

[0007] According to yet another aspect of the present disclosure, a non-transitory computer-readable medium is provided. The non-transitory computer-readable medium may include at least one set of instructions.When executed by at least one processor of a computing device, the at least one set of instructions may instruct the at least one processor to perform operations including: obtaining medical pattern images corresponding to a plurality of tracer types; determining the plurality of tracer types into a plurality of tracer clusters based on the medical pattern images, wherein each tracer cluster includes one or more tracer types of the plurality of tracer types; and for each tracer cluster, generating an image processing model corresponding to the tracer cluster by training an initial image processing model using first medical pattern images from the medical pattern images, wherein the first medical pattern images correspond to the one or more tracer types in the tracer cluster.

[0008] Additional features will be set forth in part in the description which follows, and in part will become apparent to those skilled in the art upon examination of the following and accompanying drawings, or may be learned by production or operation of the examples. The features of the present disclosure may be realized and attained by the practice or use of various aspects of the methods, instrumentalities, and combinations set forth in the detailed examples discussed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The present disclosure is further illustrated with respect to exemplary embodiments. These exemplary embodiments are described in detail with reference to the drawings. These embodiments are non-limiting exemplary embodiments in which like reference numerals represent similar structures, and wherein: Fig. 1 is a schematic diagram illustrating an imaging system according to some embodiments of the present disclosure; Fig. 2 is a block diagram of an example processing device according to some embodiments of the present disclosure; Fig. 3 is a schematic diagram illustrating an example computing device according to some embodiments of the present disclosure; Fig. 4 is a flowchart illustrating a method for generating image processing models according to some embodiments of the present disclosure; Fig. 5 is a flowchart illustrating a process of clustering a plurality of tracer types into a plurality of tracer clusters according to some embodiments of the present disclosure; Fig. 6 is a schematic diagram illustrating a process for determining a feature vector according to some embodiments of the present disclosure; Fig. 7 is a schematic diagram illustrating a training process of a first feature extraction model according to some embodiments of the present disclosure; Fig. 8 is a schematic diagram illustrating a process for determining a feature vector according to some embodiments of the present disclosure; Fig. 9 is a schematic diagram illustrating a training process of a second feature extraction model according to some embodiments of the present disclosure; Fig. 10 is a schematic diagram illustrating a process of clustering tracer types according to some embodiments of the present disclosure; Fig. 11 is a flowchart illustrating a process of processing a medical target image according to some embodiments of the present disclosure; Fig. 12 is a schematic diagram illustrating a process of processing a medical target image according to some embodiments of the present disclosure; Fig. 13 is a flowchart illustrating a process for determining a target tracer type of a medical target image according to some embodiments of the present disclosure; Fig. 14 is a schematic diagram illustrating a process for determining a first probability according to some embodiments of the present disclosure; Fig. 15 is a schematic diagram illustrating another process for determining a first probability according to some embodiments of the present disclosure; Fig. 16 is a schematic diagram illustrating a training process of a second discrimination model according to some embodiments of the present disclosure; Fig. 17 is a schematic diagram illustrating a process for determining a second probability according to some embodiments of the present disclosure; and Fig. 18 is a flowchart illustrating a process for determining whether planned tracer types need to be adjusted in accordance with some embodiments of the present disclosure. DETAILED DESCRIPTION

[0010] To illustrate the technical solutions associated with the embodiments of the present disclosure, a brief introduction to the drawings referred to in the description of the embodiments is provided below. Obviously, the drawings described below are only some examples or embodiments of the present disclosure. Those skilled in the art can apply the present disclosure to other similar scenarios according to these drawings without further creative effort. Unless apparent from the context or otherwise indicated, like reference numerals represent similar structures or operations throughout the different views of the drawings.

[0011] It is understood that the terms "system," "apparatus," "unit," and / or "module" used herein are a method of distinguishing various components, elements, parts, sections, or assemblies of different levels in ascending order. However, these terms may be replaced with other expressions if they achieve the same purpose.

[0012] As used in the disclosure and the appended claims, the singular forms "a," "an," and / or "the" may include plural forms unless the context clearly indicates otherwise. In general, the terms "comprise," "comprises," and / or "comprising," "include," "includes," and / or "including" merely request the inclusion of clearly identified steps and elements, and these steps and elements are not an exclusive list. The methods or devices may further include other steps or elements.

[0013] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art pertaining to the present disclosure. The terms used herein in the specification of the present disclosure are for the purpose of describing specific embodiments only and are not intended to be limiting of the invention. The term "and / or" as used herein includes all combinations of one or more of the relevant listed elements.

[0014] The flowcharts used in this disclosure illustrate operations that systems implement according to some embodiments of the present disclosure. It should be understood that the preceding or subsequent operations may not be performed in exact order. Instead, each step may be performed in reverse order or concurrently. Other operations may be added to these processes at any time, or a particular step or steps may be removed from these processes.

[0015] Fig. 1 is a schematic diagram illustrating an imaging system according to some embodiments of the present disclosure.

[0016] As in Fig. 1, an imaging system 100 may include a scanning device 110, a processing device 120, an end device 130, a network 140, and a storage device 150. The components of the imaging system 100 may be connected in one or more ways. For example only, as shown in Fig. 1, the scanning device 110 may be connected to the processing device 120 via the network 140. As another example, the scanning device 110 may be directly connected to the processing device 120 (as illustrated by the dashed bidirectional arrow connecting the scanning device 110 and the processing device 120).

[0017] The scanning device 110 can acquire scan data (e.g., a medical image, projection data, PET data) of a subject. In some embodiments, the subject can include a human body, organs, a body, an injury site, a tumor, a phantom, or the like. In some embodiments, the scanning device 110 can include a positron emission tomography (PET) device, a single-photon emission computed tomography (SPECT) device, a magnetic resonance imaging (MRI) device, a multimodal imaging device, or the like. In some embodiments, the scanning device 110 can scan the subject after injecting tracers of one or more tracer types into the subject to obtain a medical image of the subject.

[0018] The processing device 120 may process data and / or information obtained from the scanning device 110, the end device 130, and / or the storage device 150. For example, the processing device 120 may receive scan data from the scanning device 110 and, based on the scan data, generate a medical image (e.g., a medical template image, a medical target image, a medical reference image, etc.) corresponding to the scan data. For example, the processing device 120 may generate an image processing model based on a plurality of medical template images. As yet another example, the processing device 120 may process the medical target image based on the image processing model.In some embodiments, processing device 120 may include a central processing unit (CPU), a digital signal processor (DSP), a system on a chip (SoC), a microcontroller unit (MCU), etc., and / or any combination thereof. In some embodiments, processing device 120 may include a computer, a user console, a single server or a group of servers, etc. The group of servers may be centralized or distributed. In some embodiments, processing device 120 may be local or remote. For example, processing device 120 may access information and / or data stored in scanning device 110, end device 130, and / or storage device 150 via network 140.As another example, processing device 120 may be directly connected to scanning device 110, end device 130, and / or storage device 150 to access the stored information and / or data. In some embodiments, processing device 120 may be implemented on a cloud platform. For example only, 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 120, or a portion of processing device 120, may be integrated with scanning device 110.

[0019] End device 130 may display the medical image to a user and / or receive input from the user. For example, end device 130 displays a medical template image and a medical target image before and after processing to the user. As another example, end device 130 receives user feedback input from the user. End device 130 may include a mobile device 131, a tablet computer 132, a laptop computer 133, etc., or any combination thereof. In some embodiments, end device 130 may be part of processing device 120.

[0020] Network 140 may include any suitable network that facilitates the exchange of information and / or data for imaging system 100. In some embodiments, one or more components of imaging system 100 (e.g., scanning device 110, processing device 120, end device 130, storage device 150) may communicate information and / or data over network 140 to one or more other components of imaging system 100. In some embodiments, network 140 may include a wired network and / or a Wi-Fi network.

[0021] The storage device 150 may store data, instructions, and / or any other information. In some embodiments, the storage device 150 may store data obtained from the scanning device 110, the end device 130, and / or the processing device 120. For example, the storage device 150 may store a trained image processing model, a clustering result of a plurality of tracer types, etc. In some embodiments, the storage device 150 may include mass storage, removable storage, volatile random access memory, read-only memory (ROM), or the like, or any combination thereof. In some embodiments, the storage device 150 may be implemented on a cloud platform. In some embodiments, the storage device 150 may be connected to the network 140 to provide data to one or more other components of the imaging system 100 (e.g., the network).One or more components of the imaging system 100 may access data or instructions stored in the storage device 150 via the network 140. In some embodiments, the storage device 150 may be directly connected to or in communication with one or more other components of the imaging system 100 (e.g., the scanning device 110, the processing device 120, the storage device 150, the end device 130). In some embodiments, the storage device 150 may be part of the processing device 120.

[0022] It should be noted that the above description is provided for illustrative purposes only and is not intended to limit the scope of the present disclosure. Numerous variations and modifications may be made by those skilled in the art, guided by the contents of the present disclosure. Features, structures, methods, and other characteristics of the exemplary embodiments described herein may be combined in various ways to obtain additional and / or alternative exemplary embodiments. However, these variations and modifications do not depart from the scope of the present disclosure.

[0023] Fig. Figure 2 is a block diagram illustrating an example processing device according to some embodiments of the present disclosure. Processing device 120 may include a obtaining module 210, a determining module 220, a training module 230, a processing module 240, an adapting module 250, and an updating module 260.

[0024] The obtaining module 210 may be configured to obtain medical sample images corresponding to a variety of tracer types.

[0025] The clustering module 220 may be configured to group the plurality of tracer types into a plurality of tracer clusters based on the medical sample images.

[0026] The training module 230 may be configured to generate, for each tracer cluster, an image processing model corresponding to the tracer cluster by training an initial image processing model using first medical sample images of the medical sample images.

[0027] The processing module 240 may be configured to determine, for a target medical image corresponding to a target tracer type, whether the target tracer type is included in the plurality of tracer types; in response to determining that the target tracer type is included in the plurality of tracer types, determine one or more target tracer clusters to which the target tracer type belongs; and generate a processed target medical image by processing the target medical image using one or more image processing models corresponding to the one or more target tracer clusters.

[0028] The adaptation module 250 may be configured to, for a medical target image corresponding to a target tracer type, determine whether the target tracer type is included in the plurality of tracer types; in response to determining that the target tracer type is not included in the plurality of tracer types, determine, for each tracer cluster, a third probability that the target tracer type belongs to the tracer cluster based on the medical target image using a second discrimination model corresponding to the tracer cluster; determine one or more target tracer clusters corresponding to the target tracer type from the plurality of tracer clusters based on the third probabilities corresponding to the plurality of tracer clusters; and generate a processed medical target image by processing the medical target image using one or more image processing models corresponding to the one or more target tracer clusters.

[0029] The update module 260 may be configured to obtain medical reference images corresponding to reference tracer types not included in the plurality of tracer types; for each reference tracer type, based on the medical reference image corresponding to the reference tracer type, determine fourth probabilities that the reference tracer type belongs to the plurality of tracer clusters using second discrimination models corresponding to the plurality of tracer clusters; determine fourth position information of the reference tracer type in a feature space based on the fourth probabilities; and determine whether the plurality of tracer clusters needs to be updated based on the fourth position information of each reference tracer type.

[0030] It should be noted that the above description regarding processing device 120 is provided for illustrative purposes only and is not intended to limit the scope of the present disclosure. Various variations or modifications may be made by those of ordinary skill in the art under the guidance of the present disclosure. However, these variations and modifications do not depart from the scope of the present disclosure. For example, processing device 120 may include a memory module configured to store data generated by the above-mentioned modules of processing device 120. As yet another example, one or more modules may be integrated into a single module to perform functions thereof.

[0031] Fig. Figure 3 is a schematic diagram illustrating an example computing device according to some embodiments of the present disclosure. In some embodiments, the processing device 120 and / or the end device(s) 130 may be implemented on the computing device 300. As shown in Fig. As illustrated in Figure 3, the computing device 300 may include a display unit 310, an input device 320, a graphics processing unit (GPU) (not shown in the figure), a central processing unit (CPU) 330, a memory 340, a communication interface 350, and an input / output interface (I / O) 360. The display unit 310 is used to display information and may be a screen or a projection device. The input device 320 may be a touch layer covering the display screen, or it may be buttons, a trackball, or a touchpad installed on the housing of the computing device. It may also be an external keyboard, a touchpad, a mouse, etc. The GPU and CPU 330 are used to provide computing and control capabilities. The memory 340 includes a non-volatile storage medium and internal memory.The non-volatile storage medium stores an operating system 370 and computer programs 380. The computer programs 380 may include a browser or other suitable image processing model generation apps for receiving and rendering information related to an imaging system 100 from the processing device 120.

[0032] The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The I / O interface 360 ​​is used for information exchange between the processor and external devices. The communication interface 350 is used to communicate with external terminals in a wired or wireless manner. Radio communication can be achieved through WLAN, cellular networks, near-field communication (NFC), or other technologies. When the computer program is executed by the processor, a method for obtaining an image processing network is implemented.

[0033] In some embodiments, any other suitable component, including but not limited to a system bus or controller (not shown), may also be included in computing device 300.

[0034] To implement various modules, units, and their functionalities described in the present disclosure, computer hardware platforms may be used as the hardware platform(s) for one or more of the elements described herein. The hardware elements, operating systems, and programming languages ​​of such computers are conventional in nature, and it is assumed that those skilled in the art are sufficiently familiar with adapting these technologies to generate a high-quality image of a subject as described herein. A computer with user interface elements may be used to implement a personal computer (PC) or other type of workstation or end device, although a computer may also act as a server if appropriately programmed.It is assumed that professionals are familiar with the structure, programming and general operation of such computer equipment, and therefore the drawings should be self-explanatory.

[0035] Fig. Figure 4 is a flowchart illustrating a process for generating image processing models according to some embodiments of the present disclosure. In some embodiments, processing device 120 may perform process 400. For example, process 400 may be stored in a storage device (e.g., storage device 150, a storage unit of processing device 120) in the form of instructions, and process 400 may be implemented when processing device 120 executes the instructions.

[0036] At 410, medical sample images corresponding to a plurality of tracer types are obtained. In some embodiments, operation 410 may be performed by the obtaining module 210.

[0037] A tracer refers to a detectable and traceable marker injected into a subject during or before a medical scan (e.g., a PET scan) to obtain relevant information (e.g., biological metabolic information) about the subject. Specifically, the tracer is injected into the subject before or during the medical scan in such a way that, after reacting with a specific substance in the subject, the tracer emits rays, and the scanning device detects the rays to obtain a medical image of the subject. The tracer can mark the specific substance or tissue in the medical image to provide data to support subsequent medical analysis.For example, in metabolic studies, PET scans can be performed after a patient has been injected with fluoro-2-deoxy-D-glucose (FDG) to obtain PET images of the patient, based on which the patient's metabolism can be analyzed and studied.

[0038] With the development of medical technology, more and more tracer types are being used in medical scans. Different tracer types have different properties and can provide different information. For example, tracer types such as deoxyglucose (FDG) for imaging tumor metabolism, prostate-specific membrane antigen (PSMA) for prostate cancer diagnosis, 99m Tc-labeled methylene diphosphonates for bone imaging and 99m Include Tc-labeled diethylenetriaminepentaacetic acid for assessment of renal function.

[0039] A medical sample image refers to a historical medical image used to train image processing models. For example, a historical medical image obtained by a scanning device after a specific tracer type is injected into a subject can be used as a medical sample image. After the medical sample image is obtained in a historical scan, the physician or technician can add a tracer label, and the tracer type corresponding to the medical sample image can be determined based on the tracer label.

[0040] Different medical sample images may correspond to the same or different imaging modalities. For example, the medical sample images are all PET images. As another example, the medical sample images include PET images and MRI images. A tracer type may correspond to a plurality of medical sample images. Note that the plurality of medical sample images corresponding to a tracer type may be medical sample images obtained by scanning the same tissue from different subjects, or may be medical sample images obtained by scanning different tissues from different subjects. Certainly, the plurality of medical sample images corresponding to different tracer types may also include medical sample images of the same tissue from different subjects and / or medical sample images of different tissues from different subjects.

[0041] In some embodiments, a tracer type may have only one corresponding medical pattern image. In some embodiments, a tracer type may have multiple corresponding medical pattern images.

[0042] The processing device 120 may obtain the plurality of medical pattern images corresponding to the plurality of tracer types from a storage device. In some embodiments, the processing device 120 may obtain the plurality of medical pattern images corresponding to the plurality of tracer types from a scanning device.

[0043] At 420, a plurality of tracer clusters are determined based on the medical sample images. In some embodiments, operation 420 may be performed by the determination module 220.

[0044] In some embodiments, the plurality of tracer types are grouped into a plurality of tracer clusters based on the medical sample images. The tracer clusters refer to tracer sets obtained by grouping tracer types. Each tracer cluster contains one or more tracer types from the plurality of tracer types. One or more tracer types in the same tracer cluster have a high similarity. For example, tracer type 1 is included in tracer cluster 1, tracer types 2 and 3 are included in tracer cluster 2, and tracer types 4, 5, and 6 are included in tracer cluster 3. A tracer type may be grouped into one or more tracer clusters.

[0045] The processing device 120 can construct a feature vector to characterize each medical sample image and group the tracer types based on the feature vector to obtain the tracer clusters. Detailed descriptions of the clustering of the tracer types are provided in Fig. 5 and their associated descriptions.

[0046] At 430, for each tracer cluster, an image processing model corresponding to the tracer cluster is generated by training an initial image processing model using first medical sample images of the medical sample images. In some embodiments, operation 430 may be performed by training module 230.

[0047] Each image processing model corresponds to a tracer cluster. For example, an image processing model 1, an image processing model 2, ..., and an image processing model k may correspond to a tracer cluster 1, a tracer cluster 2, ..., and a tracer cluster k, respectively. The image processing model corresponding to a tracer cluster is used to process medical images corresponding to one or more tracer types in the tracer cluster. For example, the image processing model corresponding to tracer cluster 1 may process medical images corresponding to tracer types such as FDG, PSMA, or the like in tracer cluster 1. The processing includes image recognition, image segmentation, image enhancement (e.g., noise reduction, artifact removal, etc.), image alignment, etc.Accordingly, the image processing model includes an image recognition model, an image segmentation model, an image enhancement model, an image alignment model, or the like. One or more image processing models can be generated for a tracer cluster. Detailed descriptions regarding the processing of medical images using the image processing models are provided in . Fig. 11 and their associated descriptions.

[0048] In some embodiments, an image processing model may include one or more of a deep neural network (DNN) model, a convolutional neural network (CNN) model, a bidirectional encoder representation of transformers (BERT) model, or the like.

[0049] The first medical sample images refer to medical sample images corresponding to one or more tracer types in a single tracer cluster for training an image processing model corresponding to the tracer cluster. For example, continuing the above example, the processing device 120 may determine the PET images corresponding to the tracers FDG and PSMA in tracer cluster 1 as the first medical sample images, and these PET images are used to train the image processing model corresponding to tracer cluster 1.

[0050] The image processing model corresponding to a tracer cluster may further be generated based on first training labels corresponding to the first medical sample images. Different image processing models correspond to different first training labels. For example, for an image denoising model, the first training label of a first medical sample image is a denoising image corresponding to the first medical sample image. As another example, for an image segmentation model, the first training label of a first medical sample image is a segmented image corresponding to the first medical sample image. The first training label may be manually calibrated or determined.

[0051] During the training process, the processing device 120 may input the first medical sample images into an initial image processing model, determine a value of a loss function based on an output of the initial image processing model and the first training labels, and iteratively update the initial image processing model until an iteration condition is met. Example iteration conditions include that the value of the loss function is less than a threshold, that a difference between values ​​of the loss function in two adjacent iterations is less than a threshold, that the number of iterations exceeds a threshold, or the like.

[0052] In some embodiments of the present disclosure, the tracer types are grouped to obtain the first medical sample images corresponding to each tracer cluster, thereby training the image processing model corresponding to each tracer cluster. The scope of the image processing model corresponding to the tracer cluster is not limited to a single tracer type, but to all tracer types in the tracer cluster. At the same time, there is no need to generate a corresponding image processing model for each tracer type, which reduces the number of image processing models and the number of training samples and improves the efficiency of training the image processing models.

[0053] In some embodiments, recognizing that new tracers may continue to emerge, process 400 may further include operations 440-470 for determining whether the tracer clustering result generated above needs to be updated. Operations 440-470 may be performed by update module 260.

[0054] In 440, medical reference images corresponding to reference tracer types are obtained.

[0055] The reference tracer types refer to tracer types that are not included in the variety of tracer types discussed in operation 410. A medical reference image refers to a medical image obtained by a scanning device after injecting a reference tracer into a subject.

[0056] At 450, for each reference tracer type, fourth probabilities that the reference tracer type belongs to the plurality of tracer clusters are determined based on the medical reference image(s) corresponding to the reference tracer type(s) using second discrimination models corresponding to the plurality of tracer clusters.

[0057] The fourth probabilities of a reference tracer type refer to probabilities that the reference tracer type belongs to different tracer clusters. Determining the fourth probabilities using the second discrimination models is similar to determining the third probabilities using the second discrimination models, as described in operation 1140, and the descriptions thereof are not repeated here.

[0058] In 460, for each reference tracer type, the information of the fourth position of the reference tracer type in a feature space is determined based on the fourth probabilities.

[0059] The feature space refers to a space used to represent feature vectors corresponding to the tracer types. Further descriptions regarding the feature space are provided in operation 520 of Fig. 5 can be found.

[0060] For example, based on the fourth probabilities corresponding to the tracer clusters, distance ratio information regarding a distance between the reference tracer type and the center of each tracer cluster in the feature space can be determined. Then, based on the distance ratio information, the fourth position information of the reference tracer type in the feature space is determined. The larger the fourth probability corresponding to a tracer cluster, the shorter the distance between the reference tracer type and the center of the tracer cluster.

[0061] For example only, the fourth probabilities corresponding to tracer cluster 1, tracer cluster 2, ..., and tracer cluster k are 0.1, 0.2, ..., and 0.3, respectively, and the ratio between the distance from the reference tracer type to the center of tracer cluster 1, the center of tracer cluster 2, ..., and the center of tracer cluster k in the feature space is 10: 5: ...: 3.3. Based on the ratio and the position information of the center of each tracer cluster, the fourth position information of the reference tracer type in the feature space can be determined.

[0062] In 470, based on the fourth position information of each reference tracer type, it is determined whether the plurality of tracer clusters need to be updated.

[0063] For example, based on the fourth position information, the processing device 120 determines for each reference tracer type whether the reference tracer type can be grouped into an existing tracer cluster. If the reference tracer type is distant from the center of each tracer cluster (e.g., if the distance to the center of each tracer cluster is greater than a distance threshold), it may be determined that the reference tracer type cannot be grouped into existing tracer clusters. The processing device 120 may further determine the number of reference tracer types that cannot be grouped into the existing tracer clusters. If the number of reference tracer types exceeds a threshold, it is determined that the plurality of tracer clusters needs to be updated.As another example, based on the fourth position information of each reference tracer type and the position information of the existing tracer cluster types in the feature space, the processing device 120 determines whether the reference tracer type and the existing tracer types form a new cluster center. If a new cluster center is formed, the processing device 120 determines that the plurality of tracer clusters needs to be updated.

[0064] If it is determined that the tracer clusters need to be updated, the processing device 120 may add the reference tracer type to the tracer types described in operation 410 and add the medical reference images to the medical template images described in operation 410. The processing device 120 may repeat operation 420 to determine new tracer clusters. Furthermore, the processing device 120 may repeat operation 430 to generate new image processing models.

[0065] In some embodiments, processing device 120 may perform operations 440-470 regularly. In some embodiments, when the number of reference tracer types exceeds a threshold (ie, a certain number of new tracer types appear), processing device 120 may perform operations 440-470.

[0066] In some embodiments of the present disclosure, the tracer clusters and image processing models can be updated in a timely manner by analyzing new reference tracer types and medical reference images corresponding to the new reference tracer types. This approach enables the image processing models to be more accurate, thereby improving the reliability of medical image analysis and providing more accurate support for clinical diagnosis and research.

[0067] Fig. Figure 5 is a flowchart illustrating a process of clustering a plurality of tracer types into a plurality of tracer clusters according to some embodiments of the present disclosure. In some embodiments, process 500 may be used to implement operation 420.

[0068] In 510, feature vectors of the medical pattern images are determined by processing the medical pattern images using at least one feature extraction model.

[0069] The feature vector of a medical sample image refers to a vector of features (e.g., depth features) used to describe the medical sample image.

[0070] A feature extraction model is a machine learning model for extracting image features. The feature extraction model includes a deep neural network (DNN), a convolutional neural network (CNN), a recurrent neural network (RNN), and so on. For each medical sample image, the processing device 120 may input the medical sample image into one of at least one feature extraction model and obtain a feature vector output from the feature extraction model.

[0071] In some embodiments, the at least one feature extraction model includes a plurality of first feature extraction models. A first feature extraction model corresponds to a tracer type from the plurality of tracer types and is used to extract a feature vector of a medical specimen image corresponding to the tracer type. The feature vector for each medical specimen image is determined by processing the medical specimen image using a first feature extraction model corresponding to the same tracer type as the medical specimen image. For example, as in Fig. As shown in Figure 6, the at least one feature extraction model includes first feature extraction models 1-n (n is an integer greater than 1) corresponding to the tracer types 1-n. The medical sample images 1-n corresponding to the tracer types 1-n are respectively input to the first feature extraction models 1-n to obtain the corresponding feature vectors 1-n.

[0072] In some embodiments, a supervised learning algorithm may be used to generate a first feature extraction model corresponding to each tracer type based on second medical pattern images corresponding to each tracer type in the medical pattern image. The second medical pattern images are medical pattern images corresponding to a single tracer type.

[0073] Fig. 7 is a schematic diagram illustrating a training process of a first feature extraction model according to some embodiments of the present disclosure. As in Fig. As shown in Figure 7, a first initial model to be trained includes an encoder 750 and a decoder 760. The encoder in the present disclosure may include a CNN model, an RNN model, a transformer model, or the like, and the decoder may include an RNN model, a transformer model, or the like. Training patterns of the first initial model include a first positive pattern 710, a first negative pattern 720, a second training label 730 corresponding to the first positive pattern 710, and a second training label 740 corresponding to the first negative pattern 720. The second training label 730 of the first positive pattern 710 is 1 and the second training label 740 of the first negative pattern 720 is 0. Taking tracer type 1 (e.g., FDG) as an example, medical pattern images (e.g., FDG medical pattern images) corresponding to tracer type 1 are obtained from the medical pattern images as the first positive pattern.Other medical pattern images (e.g., PSMA medical pattern images, etc.) corresponding to other tracer types i (e.g., PSMA, etc.) (i is an integer other than 1), except for tracer type 1, are obtained as the first negative pattern. For example, the medical pattern images may be images that can reflect the metabolic distribution level of the entire body, such as chord diagrams and reconstructed diagrams generated based on original PET data. In this embodiment, no specific restrictions are imposed on the image shapes of the medical pattern images.

[0074] During the training process, the first positive pattern 710 is input to the encoder 750, which performs feature extraction to obtain a feature vector x. The feature vector x is input to the decoder 760, which outputs a probability value x of the first positive pattern 710 corresponding to tracer type 1. Similarly, the encoder 750 receives the feature vector y of the first negative pattern 720, and the decoder 760 outputs the probability value y of the first negative pattern 720 corresponding to tracer type 1. Based on the probability value x and the second training label 1, as well as the probability value y and the second training label 0, a value of a loss function is determined. The parameters of the first initial model are gradually updated by backpropagating the value of the loss function until the first initial model converges.After training is completed, the trained encoder 750 is used as the first feature extraction model corresponding to tracer type 1, and the trained decoder 760 is used as the first discrimination model corresponding to tracer type 1 (which is used in conjunction with . Fig. 14). Using the same method described above, the first feature extraction models and first discrimination models corresponding to other tracer types can be obtained separately, which will not be repeated here.

[0075] It should be noted that when training the first feature extraction model corresponding to a tracer type, the first negative pattern can be extracted from medical sample images other than the first positive pattern through uniform sampling, considering the balance between the number of positive patterns and negative patterns. For example, when training the first feature extraction model corresponding to tracer type 1, the medical sample images corresponding to tracer type 1 can be referred to as the first positive pattern, and the first negative pattern can be sampled uniformly from medical sample images corresponding to other tracer types. This type of sampling can not only satisfy the balance between the number of positive and negative samples, but also ensure the diversity and richness of the first negative pattern and improve the network training effect.

[0076] In some embodiments of the present disclosure, a first feature extraction model corresponding to each tracer type can be obtained through supervised training, allowing feature extraction to be performed separately for each tracer type, helping to improve the accuracy of feature extraction. Additionally, supervised training can learn pixel-level representations of each PET tracer, which is more suitable for situations where the tracer types in the existing PET data are comprehensive and the data volume is large.

[0077] In some embodiments, the at least one feature extraction model includes a second feature extraction model corresponding to the plurality of tracer types. The second feature extraction model is used to extract the feature vectors from medical pattern images corresponding to the plurality of tracer types. The feature vectors of the medical pattern images are determined by processing each medical pattern image using the second feature extraction model. For example, the at least one feature extraction model includes, as in Fig. 8, a second feature extraction model corresponding to the tracer types 1-n is implemented. The medical sample images 1-n corresponding to the tracer types 1-n are input to the second feature extraction model, and feature vectors 1-n corresponding to the medical sample images 1-n can be obtained.

[0078] In some embodiments, the second feature extraction model may be generated based on the medical sample image corresponding to each tracer type using an unsupervised learning algorithm.

[0079] Fig. 9 is a schematic diagram illustrating a training process of a second feature extraction model according to some embodiments of the present disclosure. As in Fig. 9, the second initial model to be trained includes an encoder 901, an encoder 902, a projector 903, and a projector 904. Projectors in the present disclosure may include a fully connected layer, etc. The structures of the encoder 901 and the encoder 902 may be the same or different. The structures of the projector 903 and the projector 904 may be the same or different. Training patterns of the second initial model may include a second positive pattern and a second negative pattern. The second positive pattern includes a pair of medical pattern images (e.g., FDG medical pattern images) corresponding to the same tracer type (e.g., FDG). The second negative pattern includes a pair of medical pattern images (e.g., FDG medical pattern image and PSMA medical pattern image) corresponding to different tracer types (e.g., FDG and PSMA).For example, with reference to . Fig. 9 the pair of medical pattern images in the second positive pattern each corresponds to tracer type 1, and the pair of medical pattern images in the second negative pattern corresponds to tracer types 1 and i, respectively.

[0080] During the training process, the two medical sample images in the second positive pattern are input to the encoder 901 and the encoder 902, respectively, and the encoder 901 and the encoder 902 output a feature vector 905 and a feature vector 906, respectively. The projector 903 and the projector 904 map the feature vector 905 and the feature vector 906 into two projection vectors 909 and 910, respectively. During the training process, the two medical sample images in the second positive pattern are input to the encoder 901 and the encoder 902, respectively, and the encoder 901 and the encoder 902 output a feature vector 907 and a feature vector 908, respectively. The projector 903 and the projector 904 map the feature vector 907 and the feature vector 908 into two projection vectors 911 and 912, respectively. Based on the projection vectors 909, 910, 911, and 912, a value for a contrast loss can be determined.Based on the contrast loss value, the parameters of the second initial model are updated until the second initial model converges. After training is complete, the trained encoder 901 can be referred to as the second feature extraction model.

[0081] In some embodiments of the present disclosure, unsupervised learning is used to train the second feature extraction model without labeling the training samples, improving training efficiency. Additionally, the unsupervised learning algorithm requires only one feature extraction model to be trained, reducing the workload involved in model training. Meanwhile, high-dimensional representations of each PET tracer can be learned by the unsupervised learning algorithm, thus possessing strong generalization capability.

[0082] In 520, the plurality of tracer types are grouped into the plurality of tracer clusters based on the feature vectors of the medical sample images.

[0083] In some embodiments, the processing device 120 uses a clustering algorithm to group the tracer types into the plurality of tracer clusters based on the feature vectors of the medical sample images corresponding to each tracer type. There is a high degree of similarity between the feature vectors corresponding to the tracer types in the same tracer cluster, while there is a large degree of variability in the feature vectors corresponding to the tracer types in different tracer clusters.

[0084] The clustering algorithm may include a soft clustering algorithm and a hard clustering algorithm. The soft clustering algorithm uses a membership function to represent the possibility or degree that each data point (i.e., a feature vector) belongs to each tracer cluster. The soft clustering algorithm may include a fuzzy C-means (FCM) algorithm, a Gaussian mixture model soft clustering algorithm (GMM soft clustering algorithm), a neural network-based soft clustering algorithm, and so on. The hard clustering algorithm directly partitions each data point (i.e., a feature vector) into a tracer cluster with the closest distance from the data point. The hard clustering algorithms may include a K-means algorithm, a K-medoids algorithm, or the like.

[0085] Fig. 10 is a schematic diagram illustrating a process of clustering tracer types according to some embodiments of the present disclosure. As shown in Fig. As shown in Figure 10, n feature vectors corresponding to n medical sample images are clustered by the K-means clustering algorithm to obtain k tracer clusters. Taking tracer cluster 1 as an example, a plurality of tracer types in tracer cluster 1 can be regarded as tracers of the same type with high similarity, and any tracer type in tracer cluster 1 and any tracer type in tracer cluster 2 can be regarded as different tracer types with high variability. Note that the number k of tracer clusters should be smaller than the number n of tracers, that is, k is smaller than n. In other words, at least one tracer cluster includes multiple tracers.

[0086] In some embodiments, a tracer type may correspond to multiple medical specimen images, each medical specimen image having a corresponding feature vector, i.e., there may be multiple feature vectors corresponding to the tracer type. During clustering, the processing device may directly group the tracer types based on the multiple feature vectors or perform certain operations on these feature vectors (e.g., after dimensionality reduction and averaging) and then group them.

[0087] In some embodiments, the clustering algorithm (e.g., the hard clustering algorithm) requires that each tracer type belong to only a single tracer cluster. In some embodiments, the clustering algorithm (e.g., the soft clustering algorithm) allows a tracer type to belong to multiple tracer clusters.

[0088] In some embodiments, step 520 may include steps 522 through 528.

[0089] In 522, a clustering result is generated by clustering the plurality of tracer types based on the feature vectors of the medical pattern images.

[0090] The clustering result can be generated based on the clustering algorithm (e.g., fuzzy C-means clustering algorithm, K-means clustering algorithm).

[0091] In 524, based on the result of the clustering, a degree of membership of each tracer type with respect to each tracer cluster is determined.

[0092] The degree of membership of a tracer type with respect to a tracer cluster refers to the probability and / or degree to which the tracer type belongs to the tracer cluster. The higher the degree of membership, the greater the probability and / or degree to which the tracer type belongs to the tracer cluster.

[0093] When clustering is performed using the soft clustering algorithm, the degree of membership of each tracer type with respect to each tracer cluster can be determined based on a membership function of the soft clustering algorithm. For example, in the case of the fuzzy C-mean clustering algorithm, the membership function can output a membership matrix U, where Uij denotes the degree to which the i-th tracer type belongs to the j-th tracer cluster. The degree of membership satisfies ∑j=1kUij=1, where k denotes the number of tracer clusters.

[0094] When clustering is performed using the hard clustering algorithm, based on the clustering result, the first position information of the cluster centers from the plurality of tracer clusters in a feature space and the second position information of the plurality of tracer types in the feature space can be determined.

[0095] The cluster centers represent central positions or central features of the corresponding tracer cluster. The first position information contains the position of the cluster center of each tracer cluster in the feature space. Based on the result of the clustering, a mean vector of feature vectors of at least one tracer type within each tracer cluster is calculated. The mean vector is mapped to the feature space to determine a position of the cluster center of the tracer cluster in the feature space, thereby obtaining the first position information. For example, the position of the cluster center of each tracer cluster in Fig. 10 represented by “X”.

[0096] The second position information includes a position of each tracer type in the feature space. Mapping the feature vector corresponding to each tracer type to the feature space allows determining the position of the tracer type in the feature space, thereby obtaining the second position information.

[0097] Furthermore, based on the first position information and the second position information, the degree of membership of each tracer type with respect to each tracer cluster is determined. Specifically, based on the first position information and the second position information, a distance between a position of each tracer type and a position of each cluster center is determined. For example, d denotes ijthe distance between the i-th tracer type and the cluster center of the j-th tracer cluster. Then, the inverse or negative exponential function of the distance is used to determine the degree of membership of the tracer type belonging to the tracer cluster. For example, U ij = 1 / d ij or U ij = epx(-βd ij ), where β denotes a fitting parameter. Finally, the multiplicity of membership degrees of the i-th tracer type is normalized so that ∑j=1kUij=1.

[0098] In 526, based on the degree of membership of each tracer type with respect to each tracer cluster, one or more first tracer types and one or more second tracer types in the plurality of tracer types are determined.

[0099] Every first tracer type belongs to one tracer cluster from the multitude of tracer clusters. Every second tracer type belongs to several tracer clusters from the multitude of tracer clusters.

[0100] For example, for each tracer type, a tracer cluster corresponding to a degree of membership greater than a threshold is determined as the tracer cluster to which it belongs. If the number of tracer clusters to which it belongs is one, the tracer type is determined as a first tracer type. If there are multiple tracer clusters to which it belongs, the tracer type is determined as a second tracer type. For example, assume, as in Fig. 10, the threshold is 0.3, the membership degrees of tracer type A belonging to tracer cluster 1, tracer cluster 2, ... tracer cluster k-1 and tracer cluster k are 0.4, 0.1, ..., 0.1, 0.1 and 0.1, respectively, then tracer type A is the first tracer type belonging to tracer cluster 1. The membership degrees of tracer type B belonging to tracer cluster 1, tracer cluster 2, ... tracer cluster k-1 and tracer cluster k are 0.3, 0.1, ..., 0.05 and 0.4, respectively, then tracer type B is the second tracer type belonging to tracer cluster 1 and tracer cluster k.

[0101] As another example, the processing device 120 may analyze the distribution of the corresponding membership degrees for each tracer type. If the largest membership degree is much greater than the others (e.g., if the difference exceeds a threshold), the tracer type is determined as the first tracer type. If there are multiple membership degrees that are greater than the others and are close to each other, the tracer type is determined as the second tracer type.

[0102] At 528, the one or more tracer types included in each tracer cluster are determined based on the one or more first tracer types and the one or more second tracer types.

[0103] Based on the tracer clusters corresponding to the first tracer type and the second tracer type, the tracer types included in each tracer cluster can be determined. For example, continuing the above example, tracer cluster 1 includes at least the first tracer type A and the second tracer type B, and tracer cluster k includes at least the second tracer type B.

[0104] In some embodiments of the present disclosure, clustering of tracer clusters is achieved by determining one or more tracer clusters to which each tracer type belongs based on the degree of membership of each tracer cluster. This type of clustering does not limit the number of tracer clusters to which each tracer cluster belongs, allowing a tracer type between two tracer clusters to be classified into two tracer clusters simultaneously, thereby increasing the flexibility of clustering.

[0105] In some embodiments of the present disclosure, feature vectors of a plurality of medical pattern images are obtained using at least one feature extraction model, and a plurality of tracer clusters are obtained by clustering different tracer types based on the feature vectors of the plurality of medical pattern images, thereby improving the accuracy of categorizing different tracer types.

[0106] Fig. 11 is a flowchart illustrating a process of processing a medical target image according to some embodiments of the present disclosure. In some embodiments, process 1100 may be performed by processing module 240.

[0107] In 1110, for a target medical image corresponding to a target tracer type, it is determined whether the target tracer type is included in the plurality of tracer types.

[0108] The medical target image refers to a medical image of a target test object to be processed. The target tracer type refers to the tracer type corresponding to the medical target image. In some embodiments, an initial tracer type of the user-input medical target image may be verified to determine the target tracer type. Detailed descriptions regarding determining the target tracer type are provided in Fig. 13 and its associated descriptions. In some embodiments, the initial tracer type entered by the user may be directly referred to as the target tracer type.

[0109] The plurality of tracer types is the plurality of tracer types that correspond to the medical sample images used to train the image processing models. Detailed descriptions regarding the plurality of tracer types can be found in the associated description of operation 410.

[0110] If it is determined that the target tracer type is included in the plurality of tracer types, the target tracer type is a known or processed tracer type, and operations 1120 through 1130 may be performed. If the target tracer type is not included in the plurality of tracer types, the target tracer type is an unknown or unprocessed tracer type, and operations 1140 through 1160 may be performed.

[0111] In 1120, one or more target tracer clusters to which the target tracer type belongs are determined.

[0112] As described in operations 420 and 520, each tracer type is grouped into one or more tracer clusters during the clustering process. The one or more tracer clusters to which the target tracer type belongs may be referred to as target tracer clusters. For example, the target tracer cluster to which target tracer type A belongs is tracer cluster 1. As another example, the target tracer cluster to which target tracer type B belongs is tracer cluster 1 and tracer cluster k.

[0113] In 1130, a processed medical target image is generated by processing the medical target image using one or more image processing models corresponding to the one or more target tracer clusters.

[0114] If the target tracer type belongs to only one target tracer cluster (i.e., the one or more target tracer clusters include a target tracer cluster), the target medical image is processed using the image processing model corresponding to the target tracer cluster to generate the processed medical target image. For example, continuing the above example, if the target tracer type is A, the target medical image is processed using the image processing model 1 corresponding to tracer cluster 1 to generate the processed medical target image.

[0115] If the target tracer type belongs to multiple target tracer clusters at the same time (i.e., the one or more target tracer clusters include multiple target tracer clusters), a processed medical target image can be generated based on the image processing models corresponding to the multiple target tracer clusters. For each target tracer cluster, the following operations can be performed. First, a membership degree of the target tracer type with respect to the target tracer cluster is determined. For example, it is determined that target tracer type B has a membership degree of 0.3 with respect to tracer cluster 1 and a membership degree of 0.4 with respect to tracer cluster k. Detailed descriptions regarding the membership degree can be found in the related descriptions at operation 520. Then, a weight value corresponding to the target tracer cluster is determined based on the membership degree.The higher the degree of membership, the higher the weight value corresponding to the degree of membership. For example, for target tracer type B, a weight value of 0.3 / (0.3+0.4)=0.43 can be determined for tracer cluster 1 and a weight value of 0.4 / (0.3+0.4)=0.57 for tracer cluster k. Furthermore, a processing result (i.e., an intermediate processing result) is generated by processing the target medical image using the image processing model corresponding to the target tracer cluster.

[0116] After generating the processing result corresponding to each target tracer cluster, the processed medical target image can be generated based on the processing result and the weight value corresponding to each tracer cluster. For example, a processing result 1 and a processing result 2 are generated by processing the medical target image using the image processing model 1 corresponding to tracer cluster 1 and the image processing model k corresponding to tracer cluster k, respectively. Then, based on the weight values ​​of 0.43 and 0.57, the processed medical target image is obtained by weighting and summing the processing result 1 and the processing result 2.

[0117] Considering that the target tracer type may belong to multiple tracer clusters, some embodiments in the present disclosure refer to the degree of membership and the weight value to secure the processing results of multiple image processing models, thereby realizing accurate processing of the medical target image. For example, the target tracer type "fibroblast activation protein (FAP)" is highly expressed in cancer-associated fibroblasts (CAFs) in most cancers, but is either weakly expressed or not expressed in normal tissues. On PET images, the target tracer FAP exhibits characteristics partially similar to those of 18F-FDG and partially similar to those of 68Ga-FAPI-46.In this case, the processing results obtained by using multiple image processing models corresponding to 18F-FDG and 68Ga-FAPI-46 are superior to those obtained by using a single image processing model. The integrin αvβ3, which is carried by the 68Ga-FAPI-RGD tracer, is a transmembrane glycoprotein that can be highly expressed in activated endothelial cells, newly formed blood vessels, and various types of tumor cells, but is weakly or not expressed in normal cells. This approach improves the processing accuracy of medical images, especially medical images corresponding to complex tracers, and provides a more reliable basis for subsequent medical image analysis, significantly enhancing the application value of medical images in clinical diagnosis and research.

[0118] In some embodiments of the present disclosure, when the target tracer type is a known or processed tracer type, the image processing model(s) corresponding to the tracer cluster(s) to which the target tracer type belongs can be directly invoked to process the target medical image, and then the processed target medical image is obtained. In this way, existing image processing models can be directly invoked to implement post-processing of the medical image and improve image processing efficiency.

[0119] In 1140, for each tracer cluster, a third probability that the target tracer type belongs to the tracer cluster is determined based on the target medical image using a second discrimination model corresponding to the tracer cluster.

[0120] The third probability indicates the probability that the medical target image corresponding to the target tracer type belongs to a specific tracer cluster.

[0121] A second discrimination model corresponds to a tracer cluster. For example, the second discrimination model 1, the second discrimination model 2, ..., and the second discrimination model k may correspond to the tracer cluster 1, the tracer cluster 2, ..., and the tracer cluster k, respectively. The second discrimination model corresponding to a tracer cluster is used to determine whether a medical image corresponds to a tracer type in the tracer cluster, i.e., the second discrimination model is used to determine whether the medical image corresponds to a tracer type in the tracer cluster. Further descriptions regarding the second discrimination model can be found in the Fig. 13 to Fig. 16 and the associated descriptions.

[0122] Specifically, the medical target image is input into a second discrimination model corresponding to each tracer cluster, and the second discrimination model outputs a corresponding third probability. If the third probability is high, the probability that the medical target image corresponds to a tracer type in the tracer cluster is high, i.e., the probability that the target tracer type belongs to the tracer cluster is high. As in Fig. 12, when the target tracer type is not included in the plurality of tracer types, the target medical image is input into the second discrimination model 1, the second discrimination model 2, ..., and the second discrimination model k, respectively, then the corresponding third probability 1, third probability 2, ..., and third probability k can be obtained, respectively.

[0123] In 1150, one or more target tracer clusters corresponding to the target tracer type are determined from the plurality of tracer clusters based on the third probabilities corresponding to the plurality of tracer clusters.

[0124] In some embodiments, processing device 120 may determine a tracer cluster corresponding to the largest third probability among the third probabilities as the target tracer cluster. In some embodiments, processing device 120 may determine one or more third probabilities that exceed a third probability threshold and determine one or more tracer clusters corresponding to these probabilities as one or more target tracer clusters.

[0125] In some embodiments, the processing device 120 may determine one or more target tracer clusters based on differences between third probabilities corresponding to the plurality of tracer clusters. Specifically, the processing device 120 may determine a difference between any two third probabilities among k third probabilities corresponding to k tracer clusters. As shown in Fig. 12, D 1,2 =|third probability 2-third probability 1|, D 2,3 =|third probability 3-third probability 2|, ...D k-1,k=| third probability k - third probability k - 1 |. The target tracer type is then determined as a first tracer type or a second tracer type based on the differences. For example, if there is a tracer cluster that has the highest third probability and the differences between the tracer cluster's third probability and the third probabilities of the other tracer clusters are all greater than or equal to a difference threshold, the target tracer type is determined as the first tracer type. If multiple tracer clusters have higher third probabilities than the other tracers and the differences in the probabilities of these tracer clusters are smaller than a difference threshold, the target tracer type is determined as the second tracer type.

[0126] Further, in response to determining that the target tracer type is the first tracer type, a target tracer cluster is determined from the plurality of tracer clusters based on the third probabilities. For example, the tracer cluster with the largest third probability is selected as the target tracer cluster. In response to determining that the target tracer type is the second tracer type, a plurality of target tracer clusters are determined from the plurality of tracer clusters based on the third probabilities. For example, a plurality of tracer clusters with third probabilities that are close to each other and higher than the third probabilities of the other tracer clusters are determined as the target tracer cluster.

[0127] In 1160, a processed medical target image is generated by processing the medical target image using one or more image processing models corresponding to the one or more target tracer clusters.

[0128] If the target tracer type corresponds to only one target tracer cluster (i.e., the one or more target tracer clusters include one target tracer cluster), the target medical image may be processed using an image processing model corresponding to the target tracer cluster to generate the processed target medical image. Detailed descriptions regarding processing the target medical image using the image processing model corresponding to the target tracer cluster can be found in the related descriptions in operation 1130 and are not repeated here.

[0129] If the target tracer type corresponds to multiple target tracer clusters (i.e., the one or more target tracer clusters include multiple target tracer clusters), the following operations are performed for each target tracer cluster. First, based on the third probability corresponding to the target tracer cluster, the weight value corresponding to the target tracer cluster is determined. The larger the third probability, the higher the weight value corresponding to the third probability. For example, the target tracer clusters include tracer cluster 2 and tracer cluster k, which correspond to third probabilities of 0.8 and 0.7, respectively. A weight value of 0.8 / (0.8+0.7)=0.53 can be determined for tracer cluster 2, and a weight value of 0.7 / (0.8+0.7)=0.47 can be determined for tracer cluster k. A processing result (i.e.an intermediately processed medical target image) is then generated by processing the medical target image using the image processing model corresponding to the target tracer cluster.

[0130] After obtaining the processing result corresponding to each target tracer cluster, the processed medical target image is generated based on the processing result and the weight value corresponding to each target tracer cluster. Detailed descriptions regarding the generation of the processed medical target image can be found in the related descriptions in operation 1130 and will not be repeated here.

[0131] In some embodiments, for an unknown or unprocessed tracer type, a medical target image corresponding to the tracer type may be input into a second discrimination model corresponding to each tracer cluster for discriminating the tracer type, so that a third probability output by the second discrimination model may be obtained. Then, one or more target tracer clusters matching the tracer type may be determined based on the third probability, so that the medical target image may be processed using the image processing model(s) corresponding to the target tracer cluster(s).In this way, medical images corresponding to a brand-new tracer type can be quickly processed using existing image processing models, thereby meeting the ever-changing needs of image processing and reducing the workload of model training.

[0132] Fig. 13 is a flowchart illustrating a process for determining a target tracer type of a medical target image according to some embodiments of the present disclosure. In some embodiments, process 1300 is used to determine the target tracer type of the medical target image described in operation 1110.

[0133] In some cases, information about the tracer, including the initial tracer type and other parameters related to image reconstruction, etc., is typically manually entered by a physician before the scan is performed. Due to manual input, it cannot be avoided that the tracer type is incorrectly entered, e.g., tracer type A is actually injected while the physician inputs tracer type B (i.e., the initial tracer type). In this case, after the target medical image is obtained through reconstruction, the processing device 120 calls an image processing model corresponding to tracer type B to process the target medical image.Because the actual target medical image is a medical image corresponding to tracer type A, adopting the image processing model corresponding to tracer type B may result in poor image processing or even failure in obtaining the processing result. To address this problem, a method for validating the tracer type input by the physician (i.e., the initial tracer type) is proposed. Fig. 13 provided.

[0134] In 1310, an initial tracer type of the target medical image input by a user is obtained.

[0135] The initial tracer type refers to the tracer type corresponding to the target medical image entered by the user.

[0136] At 1320, it is determined whether the initial tracer type is included in the plurality of tracer types. In response to determining that the initial tracer type is included in the plurality of tracer types, operations 1330 and 1341 or operations 1330 and 1342 may be performed. In response to determining that the initial tracer type is not included in the plurality of tracer types, process 1300 may be terminated or operations 1350 through 1370 may be performed.

[0137] In 1330, it is determined whether the initial tracer type is correct based on the target medical image using at least one discrimination model.

[0138] A discrimination model is a trained machine learning model for determining the tracer type corresponding to a medical image. The discrimination model may include one or a combination of one or more of the following: a support vector machine model, a logistic regression model, a deep neural network (DNN) model, a convolutional neural network (CNN) model, or the like.

[0139] In some embodiments, the at least one discrimination model includes a plurality of first discrimination models. Each first discrimination model corresponds to a tracer type from the plurality of tracer types. A first discrimination model corresponding to a tracer type can be used to discriminate whether a medical image corresponds to the tracer type or a probability that the medical image corresponds to the tracer type.

[0140] Specifically, a first probability that the medical target image corresponds to the initial tracer type is determined based on the medical target image using the first discrimination model corresponding to the initial tracer type. The first probability is the probability that the medical target image corresponds to the initial tracer type. For example, the at least one discrimination model includes, as in Fig. 14 illustrates, first discrimination models 1-n corresponding to tracer types 1-n, respectively. Assuming that the initial tracer type is tracer type 2, the target medical image or a target feature vector corresponding to the target medical image is input to the first discrimination model 2 corresponding to tracer type 2, and the first discrimination model outputs the corresponding first probability 2.

[0141] In some embodiments, the first probability may be determined by inputting the target feature vector of the medical target image corresponding to the initial tracer type into the first discrimination model. The target feature vector refers to a vector for characterizing the features of the medical target image that can be determined by the first feature extraction model corresponding to the initial tracer type. In this case, the first feature extraction model and the first discrimination model corresponding to each tracer type are jointly trained using a supervised learning algorithm based on second medical sample images corresponding to each tracer type. For example, as in Fig. 7, the first feature extraction model is the trained encoder 750 and the first discrimination model is the trained decoder 760. Detailed descriptions regarding the co-training process are in Fig. 7 and the associated descriptions.

[0142] In some embodiments, the first probability may be determined by inputting the medical target image into the first discrimination model corresponding to the initial tracer type. In such cases, the first feature extraction model and the first discrimination model corresponding to each tracer type are trained separately using the supervised learning algorithm based on second medical sample images corresponding to each tracer type. Using tracer type 2 as an example, the first feature extraction model of tracer type 2 is trained using the supervised learning algorithm described in Fig. 7, and the first discrimination model of tracer type 2 can be trained separately. Training samples of the first discrimination model include second medical sample images corresponding to tracer type 2 and its training label 1, and sample images corresponding to other tracer types and its training label 0.

[0143] In some embodiments, the at least one discrimination model includes a plurality of second discrimination models. Each second discrimination model corresponds to a tracer cluster of the plurality of tracer clusters. The second discrimination model corresponding to a tracer cluster can be used to discriminate whether a medical image corresponds to a tracer type in the tracer cluster or the probability that the medical image corresponds to a tracer type in the tracer cluster.

[0144] Specifically, one or more tracer clusters to which the initial tracer type belongs may be determined, and one or more second discrimination models corresponding to the one or more tracer clusters may be used to determine the first probability based on the target medical image. As in Fig. 15, the at least one discrimination model includes second discrimination models 1-k. Assuming that the initial tracer type is tracer type 2 belonging to tracer cluster 2, the processing device 120 may determine a first probability 2 by directly inputting the target medical image or the target feature vector of the target medical image into the second discrimination model 2 corresponding to the tracer cluster 2. The probability value output by the second discrimination model 2 represents the probability that the target medical image corresponds to a tracer type in the tracer cluster 2, which may be referred to as the first probability that the target medical image corresponds to the initial tracer type (i.e., tracer type 2).If the initial tracer type corresponds to multiple tracer clusters, the probability values ​​output by multiple second discrimination models corresponding to these tracer clusters can be averaged to determine the first probability.

[0145] In some embodiments, a second discrimination model may be obtained using the supervised learning algorithm.

[0146] As in Fig. 16, for a given tracer cluster, first medical sample images corresponding to one or more tracer types in the tracer cluster and third medical sample images corresponding to tracer types other than the tracer cluster can be determined. Furthermore, as indicated by the solid arrows in Fig. 16, the first medical sample images and the third medical sample images are used as training inputs, the tracer cluster labels of the first medical sample images and the third medical sample images can be used as training labels, and the third initial model is trained to obtain the second discrimination model corresponding to the tracer cluster. The training labels of the first medical sample images are 1 and the training labels of the third medical sample images are 0. Alternatively, as indicated by the dashed arrows in Fig. 16, feature vectors of the first medical sample images and the third medical sample images are used as training inputs. The feature vectors are determined using the second feature extraction model corresponding to the tracer clusters.

[0147] In the training process, the first medical sample images (or their feature vectors) and the third medical sample images (or their feature vectors) are input to the third initial model, and the third initial model outputs the corresponding probability value r and the probability value s, respectively. Based on the probability value r and the training label 1, as well as the probability value s and the training label 0, the value of the loss function is determined. The parameters of the third initial model are gradually updated by backpropagating the value of the loss function until the model converges and a trained second discrimination model corresponding to the tracer cluster is obtained.

[0148] In some embodiments of the present disclosure, training the second discrimination model for each tracer cluster instead of a specific tracer type may reduce the model training effort.

[0149] After determining the first probability based on the first discrimination model or the second discrimination model, it may be possible to determine whether the initial tracer type is correct based on the first probability and the first probability threshold. For example, if the first probability is greater than the first probability threshold, it can be considered that the initial tracer type entered by the user is correct; otherwise, there is a relatively high probability that the entered tracer information is incorrect.

[0150] It should be noted that when using the second discrimination model, the correct initial tracer type means that the medical target image matches the tracer cluster to which the initial tracer type belongs, and does not mean that the input initial tracer type is the same as the actually injected tracer type. However, since the medical target image matches the tracer cluster, the image processing model corresponding to the tracer cluster also matches the medical target image, ensuring the accuracy of image processing. If it is determined that the medical target image matches the tracer cluster to which the initial tracer type belongs, it can be determined that the input initial tracer type is correct.Because even if the input initial tracer type does not match the actual tracer type, it is at least ensured that the input initial tracer type is similar to the actual tracer type in its properties and belongs to the same tracer cluster, and user input errors cannot adversely affect the post-processing of the image.

[0151] In 1341, in response to determining that the initial tracer type is correct, the initial tracer type is referred to as the target tracer type.

[0152] At 1342, in response to determining that the initial tracer type is incorrect, prompt information is output indicating that the initial tracer type is incorrect, and the target tracer type is determined based on user feedback information. The prompt information is used to indicate that the initial tracer type is incorrect. The user can enter user feedback based on the prompt information. The user feedback information is the initial tracer type that the user re-enters. When the user feedback information is received, operations 1310 through 1340 can be repeated to determine the target tracer type. Alternatively, the user feedback information indicates that the initial tracer type is accurate and does not need to be changed, and the initial tracer type can be referred to as the target tracer type.

[0153] In some embodiments of the present disclosure, it is possible to verify the initial tracer type manually input by the user, thereby improving the reliability and accuracy of the input tracer information and avoiding the problem of a poor image processing effect during image post-processing due to a user input error.

[0154] In 1350, for each tracer type, a second probability that the target medical image corresponds to the tracer type is determined based on the target medical image using a first discrimination model corresponding to the tracer type.

[0155] The second probability corresponding to a tracer type is the probability that the medical target image corresponds to the tracer type. For example, as in Fig. 17, by inputting the medical target image into the first discrimination models 1-n, the corresponding second probabilities 1-n can be obtained. As another example, by inputting the target feature vector of the medical target image into the first discrimination models 1-n, the corresponding second probabilities 1-n can be obtained. Further descriptions regarding the first discrimination models are provided in Fig. 14 can be found.

[0156] In 1360, one or more tracer types whose second probabilities are greater than a second probability threshold are selected from the plurality of tracer types.

[0157] In 1370, one or more selected tracer types are output.

[0158] In some embodiments, if the initial tracer type input by the user is not a known or processed tracer type, a tracer type that better matches the target medical image (i.e., a tracer type with a second probability greater than the second probability threshold) may be determined based on the first discrimination model corresponding to each tracer type. The matching tracer type is then output to the user to assist the user in assessing and modifying the initial tracer type based on the matching tracer type, thereby improving the accuracy of the tracer information. By way of example, a tracer candidate list based on one or more tracer categories may be generated and output to the user, prompting the user to confirm the recorded tracer information based on the tracer candidate list.

[0159] Fig. 18 is a flowchart illustrating a process for determining whether planned tracer types need to be adjusted according to some embodiments of the present disclosure. In some embodiments, process 1800 may be performed by adjustment module 250. In some embodiments, process 1800 may be performed after operation 420.

[0160] In 1810, planned tracer types are determined to be injected into a target test object in order to perform a target scan of the target test object.

[0161] The target scan refers to a scan to be performed on the target test subject. The planned tracer types refer to tracer types to be injected into the target test subject before or during the target scan. In some cases, it is necessary to inject multiple tracer types to understand the condition of the target test subject from different perspectives, so that medical images corresponding to the multiple tracer types are respectively obtained by reconstruction after the target scan. However, if the multiple tracer types are related to each other (e.g., similar), it is difficult to divide image data corresponding to the multiple tracer types during image reconstruction, resulting in poor image reconstruction. Therefore, based on operations 1820-1840, it can be further determined whether the planned tracer types need to be adjusted due to high correlation.

[0162] In some embodiments, the planned tracer types to be injected into the target test object may be determined based on a scan protocol corresponding to the target scan.

[0163] In 1820, it is determined whether the planned tracer types are included in the plurality of tracer types. In response to determining that the planned tracer types are included in the plurality of tracer types, operations 1830 through 1850 are performed. In response to determining that the planned tracer types are not included in the plurality of tracer types, the current process is exited.

[0164] In 1830, a degree of correlation between the planned tracer types is determined based on the result of clustering.

[0165] Detailed descriptions regarding the result of clustering can be found in the related description in Operation 522.

[0166] The correlation degree can reflect the degree of correlation or similarity between the planned tracer types. The higher the correlation degree, the greater the similarity between the planned tracer types and the more difficult it is to distinguish the scan data corresponding to each planned tracer type during image reconstruction.

[0167] For example only, for each planned tracer type, the processing device 120 may determine a degree of membership vector of the planned tracer type to each tracer cluster based on the degree of membership of the planned tracer type with respect to each tracer cluster. The degree of membership of the planned tracer type with respect to each tracer cluster is similar to the degree of membership of the tracer type with respect to each tracer cluster in operation 524 and will not be repeated here. The degree of membership vector refers to a vector representing the degree of membership between the planned tracer type and each tracer cluster. For example, if the degrees of membership of the planned tracer type 1 with respect to the tracer clusters 1-k are B1, B2, ..., Bk, the degree of membership vector of the planned tracer type 1 is [B1, B2, ..., Bk].The processing device 120 may then determine the degree of correlation based on the membership degree vector corresponding to each scheduled tracer type. For example, the degree of correlation may include one or more degrees of correlation between one or more pairs of scheduled tracer types. The degree of correlation between a pair of scheduled tracer types may be equal to the inverse of a distance between the membership degree vectors of the pair of scheduled tracer types. The distance may include a Euclidean distance, a Manhattan distance, a cosine distance, etc. As another example, if there are multiple pairs of scheduled tracer types, their corresponding degrees of correlation may be averaged to determine the final degree of correlation.

[0168] As yet another example, processing device 120 may obtain a corresponding position of each planned tracer type in the feature space of the clustering result and determine a correlation degree between each pair of planned tracer types based on the distance between corresponding positions of the pair of planned tracer types. The closer the distance between the positions, the higher the corresponding correlation degree.

[0169] In 1840, based on the degree of correlation, it is determined whether the planned tracer types need to be adjusted. In response to determining that the planned tracer types need to be adjusted, operation 1850 may be performed. In response to determining that the planned tracer types do not need to be adjusted, the current process is terminated.

[0170] For example, if the degree of correlation between a pair of planned tracer types exceeds a correlation threshold, it is determined that the planned tracer types need to be adjusted.

[0171] In 1850, prompt information is issued indicating that the planned tracer types need to be adjusted.

[0172] In some embodiments of the present disclosure, analyzing the degree of correlation between the planned tracer types prior to scanning to determine whether to adjust the planned tracer types may prevent situations where it is difficult to distinguish scan data corresponding to the planned tracer types, thereby improving scanning accuracy and efficiency.

[0173] By describing the basic concepts, it may be quite apparent to those skilled in the art after reading this detailed disclosure that the above detailed disclosure is presented only by way of example and is not limiting. Although not expressly stated herein, those skilled in the art may make various modifications, improvements, and additions to the present disclosure. These modifications, improvements, and additions are intended to be suggested by this disclosure and are within the spirit and scope of the exemplary embodiments of the present disclosure.

[0174] In addition, certain terminology has been used to describe the embodiments of the present disclosure. For example, the terms "one embodiment," "an embodiment," and / or "some embodiments" mean that a particular feature, structure, or feature described in connection with the embodiment is included in at least one embodiment of the present disclosure. Therefore, it is emphasized and should be noted that two or more references to "an embodiment," "an embodiment," or "an alternative embodiment" in various parts of the present disclosure do not necessarily all refer to the same embodiment. In addition, some features, structures, or characteristics of one or more embodiments may properly be combined in the present disclosure.

[0175] Furthermore, the specified order of processing elements or sequences, or the use of numbers, letters, or other designations, is therefore not intended to limit the claimed processes and methods to any order unless specified in the claims. Although the above disclosure discusses some embodiments of the invention that are presently believed to be useful through various examples, it should be understood that such details are for illustrative purposes only, and the additional claims are not limited to the disclosed embodiments. Instead, the claims are intended to cover all combinations of corrections and equivalents consistent with the spirit and scope of the embodiments of the present disclosure.For example, although the implementation of various components described above may be embedded in a hardware device, it may also be implemented as a pure software solution, e.g., as an installation on an existing server or mobile device.

[0176] It should also be noted that in the foregoing description of the embodiments of the present disclosure, various features are sometimes grouped into a single embodiment, figure, or description thereof to streamline the disclosure and aid in understanding one or more of the various embodiments. However, this disclosure does not imply that the objective of the present disclosure requires more features than those recited in the claims. Rather, the claimed subject matter may reside in fewer than all features of a single previously disclosed embodiment.

[0177] In some embodiments, numbers expressing quantities or properties used to describe and claim particular embodiments of the present disclosure may, in some cases, be understood by the term "about," "approximately," or "substantially." For example, "about," "approximately," or "substantially" may indicate a deviation of ±20% of the described value, unless otherwise noted. Accordingly, the numerical parameters set forth in the written description and the appended claims are, in some embodiments, approximate values ​​that may vary depending on the desired properties to be obtained by a particular embodiment. In some embodiments, the numerical parameters should be interpreted in terms of the number of significant figures provided and by applying common rounding techniques.Notwithstanding that the numerical ranges and parameters setting forth the broad scope of some embodiments of the present disclosure are approximate, the numerical values ​​set forth in the specific examples are presented as accurately as possible.

[0178] Each of the patents, patent applications, publications of patent applications, and other materials, such as articles, books, specifications, publications, documents, things, and / or the like, referred to herein are hereby incorporated by reference in their entirety for all purposes. Historical application documents that are inconsistent or contradictory with the contents of the present disclosure, as well as documents (presently or later attached to the present specification) that limit the broadest scope of the claims of the present disclosure, are excluded.For example, if there is any inconsistency or conflict between the description, definition and / or use of a term associated with any of the Integrated Materials and that associated with this document, the description, definition and / or use of the term in this document shall prevail.

[0179] Finally, it should be understood that the embodiments of the present disclosure disclosed herein illustrate the principles of the embodiments of the present disclosure. Other modifications that may be used may fall within the scope of the present disclosure. Thus, for example, but not limited to, alternative configurations of the embodiments of the present disclosure may be utilized in accordance with the guidance contained herein. Accordingly, the embodiments of the present disclosure are not limited to those precisely shown and described. QUOTES CONTAINED IN THE DESCRIPTION

[0000] This list of documents submitted by the applicant was generated automatically and is included solely for the convenience of the reader. This list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions. Cited patent literature

[0000] CN 202410424881.3

[0001]

Claims

[1] A method for generating image processing models, implemented on a computing device having at least one processor and at least one memory device, the method comprising: Obtaining medical sample images corresponding to a variety of tracer types; Determining a plurality of tracer clusters based on the medical sample images, wherein each tracer cluster includes one or more tracer types of the plurality of tracer types; and for each tracer cluster, generating an image processing model corresponding to the tracer cluster by training an initial image processing model using first medical sample images of the medical sample images, wherein the first medical sample images correspond to the one or more tracer types in the tracer cluster. [2] The method of claim 1, wherein determining a plurality of tracer clusters based on the medical pattern images comprises: Determining feature vectors of the medical pattern images by processing the medical pattern images using at least one feature extraction model, wherein the at least one feature extraction model is at least one trained machine learning model; and Clustering the plurality of tracer types into the plurality of tracer clusters based on the feature vectors of the medical sample images. [3] The method of claim 2, wherein determining the plurality of tracer clusters based on the feature vectors of the medical pattern images comprises: Generating a clustering result by clustering the plurality of tracer types based on the feature vectors of the medical pattern images; Determining a degree of membership of each tracer type with respect to each tracer cluster based on the result of the clustering; Determining one or more first tracer types and one or more second tracer types in the plurality of tracer types based on the degree of membership of each tracer type with respect to each tracer cluster, wherein each first tracer type belongs to one tracer cluster of the plurality of tracer clusters and each second tracer type belongs to multiple tracer clusters of the plurality of tracer clusters; and Determining the one or more first tracer types included in each tracer cluster based on the one or more first tracer types and the one or more second tracer types. [4] The method of claim 1, wherein the method further comprises: for a target medical image corresponding to a target tracer type, determining whether the target tracer type is included in the plurality of tracer types; in response to determining that the target tracer type is included in the plurality of tracer types, determining one or more target tracer clusters to which the target tracer type belongs; and Generating a processed medical target image by processing the medical target image using one or more image processing models corresponding to the one or more tracer clusters. [5] The method of claim 4, wherein the one or more target tracer clusters include a plurality of target tracer clusters and the processed medical target image is generated by: for each target tracer cluster, Determining a degree of membership of the target tracer type with respect to the target tracer cluster; Determining a weight value corresponding to the target tracer cluster based on the degree of membership; and Generating a processing result by processing the target medical image using the image processing model corresponding to the target tracer cluster; and Generate the processed target medical image based on the processing result and the weight value corresponding to each target tracer cluster. [6] The method of claim 4, wherein the target tracer type of the target medical image is determined by: Obtaining a first tracer type of the target medical image input by a user; in response to determining that the initial tracer type is included in the plurality of tracer types, Determining whether the initial tracer type is correct based on the target medical image using at least one discrimination model, wherein the at least one discrimination model is at least one trained machine learning model; in response to determining that the initial tracer type is correct, designating the initial tracer type as the target tracer type; or in response to determining that the initial tracer type is incorrect, outputting prompt information indicating that the initial tracer type is incorrect and determining the target tracer type based on user feedback information. [7] The method of claim 6, wherein the at least one discrimination model includes a plurality of first discrimination models, each corresponding to one of the plurality of tracer types, wherein determining whether the initial tracer type is correct comprises: Determining a first probability that the medical target image corresponds to the initial tracer type based on the medical target image using the first discrimination model corresponding to the initial tracer type; and Determine whether the initial tracer type is correct based on the first probability and a first probability threshold. [8] The method of claim 6, wherein the at least one discrimination model includes a plurality of second discrimination models, each corresponding to one of the plurality of tracer clusters, wherein determining whether the initial tracer type is correct comprises: Determining one or more tracer clusters to which the initial tracer type belongs; Determining a first probability that the target medical image corresponds to the initial tracer type based on the target medical image using the one or more second discrimination models corresponding to the one or more tracer clusters; Determine whether the initial tracer type is correct based on the first probability and a first probability threshold. [9] The method of claim 4, wherein the target tracer type of the target medical image is determined by: Obtaining a first tracer type of the target medical image input by a user; in response to determining that the initial tracer type is not included in the plurality of tracer types, for each tracer type, determining a second probability that the target medical image corresponds to the tracer type based on the target medical image using a first discrimination model corresponding to the tracer type; Selecting one or more tracer types from the plurality of tracer types whose second probabilities are greater than a second probability threshold; and Output the one or more selected tracer types. [10] The method of claim 1, wherein the method further comprises: for a target medical image corresponding to a target tracer type, determining whether the target tracer type is included in the plurality of tracer types; in response to determining that the target tracer type is not included in the plurality of tracer types, for each tracer cluster, determining a third probability that the target tracer type belongs to the tracer cluster based on the target medical image using a second discrimination model corresponding to the tracer cluster; Determining one or more target tracer clusters corresponding to the target tracer type from the plurality of tracer clusters based on the third probabilities corresponding to the plurality of tracer clusters; and Generating a processed medical target image by processing the medical target image using one or more image processing models corresponding to the one or more tracer clusters.

Images